fix(ai): resolve UAF and require provider argument instead of defaults

This enforces explicit provider specification. Automatic fallback to preferences or "openai" is removed. Users must now pass the provider name explicitly. default provider functionality was improperly removed and implemented (some parts still remain intact), leading to UAF.
fix(ai): properly decode \uXXXX JSON escape sequences as UTF-8
2026-05-19 16:42:22 +00:00 · 2026-05-19 16:40:21 +00:00 · 2026-05-16 15:47:49 +00:00 · 2026-05-16 15:33:42 +00:00 · 2026-05-15 14:14:11 +00:00 · 2026-05-15 11:58:25 +00:00
37 changed files with 5714 additions and 124 deletions
--- a/.gitignore
+++ b/.gitignore
@@ -17,6 +17,8 @@ compile_commands.json
 .project
 .settings/
 .vscode/
+.kilo/
+.roo/
 *.plist/

 # autotools
--- a/Makefile.am
+++ b/Makefile.am
@@ -37,6 +37,7 @@ core_sources = \
 	src/ui/window_list.c src/ui/window_list.h \
 	src/ui/rosterwin.c src/ui/occupantswin.c \
 	src/ui/buffer.c src/ui/buffer.h \
+	src/ui/aiwin.c \
 	src/ui/chatwin.c \
 	src/ui/mucwin.c \
 	src/ui/privwin.c \
@@ -78,7 +79,8 @@ core_sources = \
 	src/plugins/themes.c src/plugins/themes.h \
 	src/plugins/settings.c src/plugins/settings.h \
 	src/plugins/disco.c src/plugins/disco.h \
-	src/ui/tray.h src/ui/tray.c
+	src/ui/tray.h src/ui/tray.c \
+	src/ai/ai_client.h src/ai/ai_client.c

 unittest_sources = \
 	src/xmpp/contact.c src/xmpp/contact.h src/common.c \
@@ -89,6 +91,7 @@ unittest_sources = \
 	src/xmpp/chat_state.h src/xmpp/chat_state.c \
 	src/xmpp/roster_list.c src/xmpp/roster_list.h \
 	src/xmpp/xmpp.h src/xmpp/form.c \
+	src/ai/ai_client.h src/ai/ai_client.c \
 	src/ui/ui.h \
 	src/otr/otr.h \
 	src/pgp/gpg.h \
@@ -138,6 +141,7 @@ unittest_sources = \
 	tests/unittests/xmpp/stub_message.c \
 	tests/unittests/ui/stub_ui.c tests/unittests/ui/stub_ui.h \
 	tests/unittests/ui/stub_vcardwin.c \
+	tests/unittests/ui/stub_ai.c \
 	tests/unittests/log/stub_log.c \
 	tests/unittests/chatlog/stub_chatlog.c \
 	tests/unittests/database/stub_database.c \
@@ -174,6 +178,7 @@ unittest_sources = \
 	tests/unittests/test_callbacks.c tests/unittests/test_callbacks.h \
 	tests/unittests/test_plugins_disco.c tests/unittests/test_plugins_disco.h \
 	tests/unittests/test_forced_encryption.c tests/unittests/test_forced_encryption.h \
+	tests/unittests/test_ai_client.c tests/unittests/test_ai_client.h \
 	tests/unittests/test_database_export.c tests/unittests/test_database_export.h \
 	tests/unittests/test_database_stress.c tests/unittests/test_database_stress.h \
 	tests/unittests/unittests.c
@@ -198,6 +203,7 @@ functionaltest_sources = \
 	tests/functionaltests/test_autoping.c tests/functionaltests/test_autoping.h \
 	tests/functionaltests/test_disco.c tests/functionaltests/test_disco.h \
 	tests/functionaltests/test_export_import.c tests/functionaltests/test_export_import.h \
+	tests/functionaltests/test_ai.c tests/functionaltests/test_ai.h \
 	tests/functionaltests/functionaltests.c

 main_source = src/main.c
--- a/src/ai/ai_client.c
+++ b/src/ai/ai_client.c
--- a/src/ai/ai_client.h
+++ b/src/ai/ai_client.h
@@ -0,0 +1,291 @@
+#ifndef AI_CLIENT_H
+#define AI_CLIENT_H
+
+#include <glib.h>
+
+/**
+ * @brief AI message structure for conversation history.
+ */
+typedef struct ai_message_t
+{
+    gchar* role;    /* "user" or "assistant" */
+    gchar* content; /* Message content */
+} AIMessage;
+
+/**
+ * @brief AI provider configuration.
+ */
+typedef struct ai_provider_t
+{
+    gchar* name;           /* Provider name (e.g., "openai", "perplexity") */
+    gchar* api_url;        /* API endpoint URL */
+    gchar* project_id;     /* Optional project ID (for some providers) */
+    gchar* default_model;  /* Default model for this provider */
+    GHashTable* settings;  /* Extensible per-provider settings (e.g., tools=enabled, search=disabled) */
+    GList* models;         /* Cached models (gchar*) */
+    gboolean models_fresh; /* Whether models cache is current */
+    gint32 ref_count;      /* Reference count (atomic) */
+} AIProvider;
+
+/**
+ * @brief AI chat session structure.
+ */
+typedef struct ai_session_t
+{
+    pthread_mutex_t lock; /* Protects all session fields below */
+    gchar* provider_name; /* Provider name */
+    AIProvider* provider; /* Provider configuration */
+    gchar* model;         /* Model identifier (e.g., "gpt-4", "sonar") */
+    gchar* api_key;       /* API key for this session */
+    GList* history;       /* Conversation history (GList of AIMessage*) */
+    gint32 ref_count;     /* Reference count (atomic) */
+} AISession;
+
+/* ========================================================================
+ * Provider Management
+ * ======================================================================== */
+
+/**
+ * Initialize the AI client and load default providers.
+ */
+void ai_client_init(void);
+
+/**
+ * Shutdown the AI client and free resources.
+ */
+void ai_client_shutdown(void);
+
+/**
+ * Get a provider by name.
+ * @param name The provider name (e.g., "openai", "perplexity")
+ * @return AIProvider*, or NULL if not found
+ */
+AIProvider* ai_get_provider(const gchar* name);
+
+/**
+ * Add or update a provider configuration.
+ * @param name The provider name
+ * @param api_url The API endpoint URL
+ * @return New AIProvider* (caller must unref when done)
+ */
+AIProvider* ai_add_provider(const gchar* name, const gchar* api_url);
+
+/**
+ * Remove a provider by name.
+ * @param name The provider name
+ * @return TRUE if provider was removed, FALSE if not found
+ */
+gboolean ai_remove_provider(const gchar* name);
+
+/**
+ * Increment the reference count of a provider.
+ * @param provider The provider to reference
+ * @return The same provider pointer
+ */
+AIProvider* ai_provider_ref(AIProvider* provider);
+
+/**
+ * Decrement the reference count of a provider.
+ * @param provider The provider to unreference
+ */
+void ai_provider_unref(AIProvider* provider);
+
+/**
+ * List all configured providers.
+ * @return GList of AIProvider* (caller must not free the list or providers,
+ *         and must not unref the returned providers)
+ */
+GList* ai_list_providers(void);
+
+/**
+ * Find a provider name for autocomplete.
+ * @param search_str The search string
+ * @param previous Whether to go to previous match
+ * @param context Unused
+ * @return Provider name, or NULL if not found
+ */
+gchar* ai_providers_find(const char* const search_str, gboolean previous, void* context);
+
+/**
+ * Reset the provider autocomplete state.
+ * Called from cmd_ac_reset() to clear autocomplete state.
+ */
+void ai_providers_reset_ac(void);
+
+/**
+ * Get the API key for a provider.
+ * @param provider_name The provider name
+ * @return The API key, or NULL if not set (caller must free)
+ */
+gchar* ai_get_provider_key(const gchar* provider_name);
+
+/**
+ * Set the API key for a provider.
+ * @param provider_name The provider name
+ * @param api_key The API key to set
+ */
+void ai_set_provider_key(const gchar* provider_name, const gchar* api_key);
+
+/**
+ * Set the default model for a provider.
+ * @param provider_name The provider name
+ * @param model The default model name
+ */
+void ai_set_provider_default_model(const gchar* provider_name, const gchar* model);
+
+/**
+ * Get the default model for a provider.
+ * @param provider_name The provider name
+ * @return The default model name (caller must not free), or NULL if not set
+ */
+const gchar* ai_get_provider_default_model(const gchar* provider_name);
+
+/**
+ * Set a custom setting for a provider.
+ * @param provider_name The provider name
+ * @param setting The setting key (e.g., "tools", "search")
+ * @param value The setting value
+ */
+void ai_set_provider_setting(const gchar* provider_name, const gchar* setting, const gchar* value);
+
+/**
+ * Get a custom setting for a provider.
+ * @param provider_name The provider name
+ * @param setting The setting key
+ * @return The setting value (caller must free), or NULL if not set
+ */
+gchar* ai_get_provider_setting(const gchar* provider_name, const gchar* setting);
+
+/**
+ * Fetch available models from a provider's API.
+ * @param provider_name The provider name
+ * @param user_data User data (ProfAiWin* for UI display)
+ * @return TRUE if the request was successfully queued, FALSE otherwise
+ */
+gboolean ai_fetch_models(const gchar* provider_name, gpointer user_data);
+
+/**
+ * Check if models cache is fresh for a provider.
+ * @param provider_name The provider name
+ * @return TRUE if models are fresh, FALSE otherwise
+ */
+gboolean ai_models_are_fresh(const gchar* provider_name);
+
+/* ========================================================================
+ * Parsing helpers (exposed for testing)
+ * ======================================================================== */
+
+/**
+ * Parse model IDs from an OpenAI-compatible API response.
+ * Expected format: {"object":"list","data":[{"id":"model1",...},...]}
+ * @param provider The provider to add models to
+ * @param json The JSON response from the API
+ */
+void ai_parse_models_from_json(AIProvider* provider, const gchar* json);
+
+/**
+ * Extract assistant content from an LLM response body. Tries Perplexity
+ * /v1/responses "text" first, falls back to OpenAI "content". Decodes
+ * the \" and \n escape sequences only.
+ * @param response_json The JSON body from the provider
+ * @return Newly allocated content string, or NULL if not found (caller frees)
+ */
+gchar* ai_parse_response(const gchar* response_json);
+
+/**
+ * Extract human-readable error message from an API error envelope.
+ * Expected format: {"error":{"message":"...","type":"...","code":...}}
+ * Decodes \" \n \\ \t escapes.
+ * @param error_json The JSON body of the error response
+ * @return Newly allocated error message, or NULL if not parseable (caller frees)
+ */
+gchar* ai_parse_error_message(const gchar* error_json);
+
+/* ========================================================================
+ * Session Management
+ * ======================================================================== */
+
+/**
+ * Create a new AI session with the specified provider and model.
+ * @param provider_name The provider name (e.g., "openai")
+ * @param model The model identifier (e.g., "gpt-4")
+ * @return New AISession*, or NULL on failure
+ */
+AISession* ai_session_create(const gchar* provider_name, const gchar* model);
+
+/**
+ * Increment the reference count of an AI session.
+ * @param session The session to reference
+ * @return The same session pointer
+ */
+AISession* ai_session_ref(AISession* session);
+
+/**
+ * Decrement the reference count and free the session when it reaches zero.
+ * @param session The session to unreference
+ */
+void ai_session_unref(AISession* session);
+
+/**
+ * Add a message to the session history.
+ * @param session The session
+ * @param role The message role ("user" or "assistant")
+ * @param content The message content
+ */
+void ai_session_add_message(AISession* session, const gchar* role, const gchar* content);
+
+/**
+ * Clear the conversation history.
+ * @param session The session
+ */
+void ai_session_clear_history(AISession* session);
+
+/**
+ * Get the current model for a session.
+ * @param session The session
+ * @return The model name (caller must not free)
+ */
+const gchar* ai_session_get_model(AISession* session);
+
+/**
+ * Set the model for a session.
+ * @param session The session
+ * @param model The model name
+ */
+void ai_session_set_model(AISession* session, const gchar* model);
+
+/**
+ * Atomically switch session provider, model, and API key.
+ * All mutations happen under the session lock to prevent races with
+ * _ai_request_thread() which snapshots session state before making requests.
+ *
+ * @param session The session
+ * @param provider_name New provider name
+ * @param model New model identifier
+ * @param api_key New API key (caller must free after calling this)
+ */
+void ai_session_switch(AISession* session, const gchar* provider_name,
+                       const gchar* model, gchar* api_key);
+
+/* ========================================================================
+ * Request Handling
+ * ======================================================================== */
+
+/**
+ * Send a prompt to the AI provider asynchronously.
+ * @param session The AI session containing provider and model
+ * @param prompt The prompt to send
+ * @param user_data User data (ProfAiWin* for UI display)
+ * @return TRUE if the request was successfully queued, FALSE otherwise
+ */
+gboolean ai_send_prompt(AISession* session, const gchar* prompt,
+                        gpointer user_data);
+
+/**
+ * Escape a string for JSON embedding.
+ * @param str The string to escape
+ * @return Newly allocated escaped string (caller must free)
+ */
+gchar* ai_json_escape(const gchar* str);
+
+#endif /* AI_CLIENT_H */
--- a/src/command/cmd_ac.c
+++ b/src/command/cmd_ac.c
@@ -66,6 +66,8 @@
 #include "omemo/omemo.h"
 #endif

+#include "ai/ai_client.h"
+
 static char* _sub_autocomplete(ProfWin* window, const char* const input, gboolean previous);
 static char* _notify_autocomplete(ProfWin* window, const char* const input, gboolean previous);
 static char* _theme_autocomplete(ProfWin* window, const char* const input, gboolean previous);
@@ -137,6 +139,7 @@ static char* _strophe_autocomplete(ProfWin* window, const char* const input, gbo
 static char* _adhoc_cmd_autocomplete(ProfWin* window, const char* const input, gboolean previous);
 static char* _vcard_autocomplete(ProfWin* window, const char* const input, gboolean previous);
 static char* _force_encryption_autocomplete(ProfWin* window, const char* const input, gboolean previous);
+static char* _ai_autocomplete(ProfWin* window, const char* const input, gboolean previous);

 static char* _script_autocomplete_func(const char* const prefix, gboolean previous, void* context);

@@ -280,11 +283,16 @@ static Autocomplete history_switch_ac;
 static Autocomplete color_ac;
 static Autocomplete correction_ac;
 static Autocomplete avatar_ac;
+static Autocomplete ai_subcommands_ac;
+static Autocomplete ai_set_subcommands_ac;
+static Autocomplete ai_set_custom_subcommands_ac;
+static Autocomplete ai_remove_subcommands_ac;
 static Autocomplete url_ac;
 static Autocomplete executable_ac;
 static Autocomplete executable_param_ac;
 static Autocomplete intype_ac;
 static Autocomplete mood_ac;
+static Autocomplete ai_models_ac;
 static Autocomplete mood_type_ac;
 static Autocomplete strophe_ac;
 static Autocomplete strophe_sm_ac;
@@ -456,7 +464,12 @@ static Autocomplete* all_acs[] = {
    &vcard_togglable_param_ac,
    &vcard_address_type_ac,
    &force_encryption_ac,
-    &force_encryption_policy_ac
+    &force_encryption_policy_ac,
+    &ai_subcommands_ac,
+    &ai_set_subcommands_ac,
+    &ai_set_custom_subcommands_ac,
+    &ai_remove_subcommands_ac,
+    &ai_models_ac
 };

 static GHashTable* ac_funcs = NULL;
@@ -1119,6 +1132,7 @@ cmd_ac_init(void)
    autocomplete_add(statusbar_show_ac, "name");
    autocomplete_add(statusbar_show_ac, "number");
    autocomplete_add(statusbar_show_ac, "read");
+    autocomplete_add(statusbar_show_ac, "dbbackend");

    autocomplete_add(statusbar_tabmode_ac, "actlist");
    autocomplete_add(statusbar_tabmode_ac, "dynamic");
@@ -1133,9 +1147,6 @@ cmd_ac_init(void)
    autocomplete_add(status_state_ac, "xa");
    autocomplete_add(status_state_ac, "dnd");

-    autocomplete_add(logging_ac, "chat");
-    autocomplete_add(logging_ac, "group");
-
    autocomplete_add(privacy_ac, "logging");
    autocomplete_add(privacy_ac, "os");

@@ -1169,6 +1180,26 @@ cmd_ac_init(void)
    autocomplete_add(correction_ac, "off");
    autocomplete_add(correction_ac, "char");

+    autocomplete_add_unsorted(ai_subcommands_ac, "set", FALSE);
+    autocomplete_add_unsorted(ai_subcommands_ac, "remove", FALSE);
+    autocomplete_add_unsorted(ai_subcommands_ac, "start", FALSE);
+    autocomplete_add_unsorted(ai_subcommands_ac, "clear", FALSE);
+    autocomplete_add_unsorted(ai_subcommands_ac, "providers", FALSE);
+    autocomplete_add_unsorted(ai_subcommands_ac, "switch", FALSE);
+    autocomplete_add_unsorted(ai_subcommands_ac, "models", FALSE);
+
+    autocomplete_add_unsorted(ai_set_subcommands_ac, "provider", FALSE);
+    autocomplete_add_unsorted(ai_set_subcommands_ac, "token", FALSE);
+    autocomplete_add_unsorted(ai_set_subcommands_ac, "default-model", FALSE);
+    autocomplete_add_unsorted(ai_set_subcommands_ac, "custom", FALSE);
+
+    autocomplete_add_unsorted(ai_set_custom_subcommands_ac, "tools", FALSE);
+    autocomplete_add_unsorted(ai_set_custom_subcommands_ac, "search", FALSE);
+    autocomplete_add_unsorted(ai_set_custom_subcommands_ac, "memory", FALSE);
+    autocomplete_add_unsorted(ai_set_custom_subcommands_ac, "plugins", FALSE);
+
+    autocomplete_add_unsorted(ai_remove_subcommands_ac, "provider", FALSE);
+
    autocomplete_add(avatar_ac, "set");
    autocomplete_add(avatar_ac, "disable");
    autocomplete_add(avatar_ac, "get");
@@ -1444,6 +1475,7 @@ cmd_ac_init(void)
    g_hash_table_insert(ac_funcs, "/wins", _wins_autocomplete);
    g_hash_table_insert(ac_funcs, "/wintitle", _wintitle_autocomplete);
    g_hash_table_insert(ac_funcs, "/force-encryption", _force_encryption_autocomplete);
+    g_hash_table_insert(ac_funcs, "/ai", _ai_autocomplete);
 }

 void
@@ -1662,6 +1694,7 @@ cmd_ac_reset(ProfWin* window)
    win_reset_search_attempts();
    win_close_reset_search_attempts();
    plugins_reset_autocomplete();
+    ai_providers_reset_ac();
 }

 void
@@ -4044,11 +4077,6 @@ _logging_autocomplete(ProfWin* window, const char* const input, gboolean previou
        return result;
    }

-    result = autocomplete_param_with_func(input, "/logging chat", prefs_autocomplete_boolean_choice, previous, NULL);
-    if (result) {
-        return result;
-    }
-
    result = autocomplete_param_with_ac(input, "/logging group", logging_group_ac, TRUE, previous);
    return result;
 }
@@ -4450,4 +4478,163 @@ _force_encryption_autocomplete(ProfWin* window, const char* const input, gboolea

    result = autocomplete_param_with_ac(input, "/force-encryption", force_encryption_ac, TRUE, previous);
    return result;
-}
+}
+
+/* Forward declaration */
+static char* _ai_provider_and_model_autocomplete(ProfWin* window, const char* const input,
+                                                 gboolean previous, const char* cmd);
+
+static char*
+_ai_autocomplete(ProfWin* window, const char* const input, gboolean previous)
+{
+    char* result = NULL;
+
+    /* Parse once for reuse - /ai <subcommand> [<arg1>] [<arg2>] */
+    gboolean parse_result = FALSE;
+    auto_gcharv gchar** args = parse_args(input, 1, 4, &parse_result);
+    int num_args = g_strv_length(args);
+
+    /* Top-level /ai <subcommand> autocomplete (e.g., /ai s<tab> -> /ai set) */
+    result = autocomplete_param_with_func(input, "/ai set provider", ai_providers_find, previous, NULL);
+    if (result) {
+        return result;
+    }
+
+    // /ai set token <provider> <token> - autocomplete provider names
+    result = autocomplete_param_with_func(input, "/ai set token", ai_providers_find, previous, NULL);
+    if (result) {
+        return result;
+    }
+
+    // /ai set default-model <provider> <model> - autocomplete provider names
+    result = autocomplete_param_with_func(input, "/ai set default-model", ai_providers_find, previous, NULL);
+    if (result) {
+        return result;
+    }
+
+    // /ai set custom <provider> <setting> <value> - autocomplete provider names
+    result = autocomplete_param_with_func(input, "/ai set custom", ai_providers_find, previous, NULL);
+    if (result) {
+        return result;
+    }
+
+    // /ai set custom <provider> <setting> - autocomplete settings
+    /* args[0]="set", args[1]="custom", args[2]=provider (if typed), args[3]=setting (if typed) */
+    if (num_args == 3 && g_strcmp0(args[1], "custom") == 0) {
+        /* /ai set custom <provider> - check if provider is valid */
+        if (ai_get_provider(args[2])) {
+            /* Valid provider, try settings autocomplete */
+            result = autocomplete_param_with_ac(input, "/ai set custom ", ai_set_custom_subcommands_ac, TRUE, previous);
+            if (result) {
+                return result;
+            }
+        }
+    }
+    // /ai set <subcommand> - autocomplete subcommands
+    result = autocomplete_param_with_ac(input, "/ai set", ai_set_subcommands_ac, TRUE, previous);
+    if (result) {
+        return result;
+    }
+
+    // /ai remove provider <name> - autocomplete provider names
+    result = autocomplete_param_with_func(input, "/ai remove provider", ai_providers_find, previous, NULL);
+    if (result) {
+        return result;
+    }
+
+    result = autocomplete_param_with_ac(input, "/ai remove", ai_remove_subcommands_ac, TRUE, previous);
+    if (result) {
+        return result;
+    }
+
+    // /ai start <provider> <model> - autocomplete provider names and model names
+    result = _ai_provider_and_model_autocomplete(window, input, previous, "/ai start");
+    if (result) {
+        return result;
+    }
+
+    // /ai switch <provider> <model> - autocomplete provider names and model names
+    result = _ai_provider_and_model_autocomplete(window, input, previous, "/ai switch");
+    if (result) {
+        return result;
+    }
+
+    // /ai set default-model <provider> <model> - autocomplete provider names and model names
+    result = _ai_provider_and_model_autocomplete(window, input, previous, "/ai set default-model");
+    if (result) {
+        return result;
+    }
+
+    // /ai models <provider> - autocomplete provider names
+    result = autocomplete_param_with_func(input, "/ai models", ai_providers_find, previous, NULL);
+    if (result) {
+        return result;
+    }
+
+    // /ai clear - no autocomplete
+    // /ai providers - no autocomplete
+
+    result = autocomplete_param_with_ac(input, "/ai", ai_subcommands_ac, FALSE, previous);
+
+    return result;
+}
+
+/**
+ * Autocomplete provider names and model names for /ai <cmd> <provider> <model> patterns.
+ * First tries to autocomplete provider names, then model names if provider is valid.
+ * Uses static ai_models_ac to preserve cycling state (last_found) across calls.
+ * The cmd_prefix should NOT include trailing space (e.g., "/ai start" not "/ai start ").
+ */
+static char*
+_ai_provider_and_model_autocomplete(ProfWin* window, const char* const input, gboolean previous, const char* cmd)
+{
+    char* result;
+
+    /* First, try provider name autocomplete */
+    result = autocomplete_param_with_func(input, cmd, ai_providers_find, previous, NULL);
+    if (result) {
+        return result;
+    }
+
+    /* Provider was specified, try model name autocomplete */
+    /* Build the full command prefix with space */
+    auto_gchar gchar* cmd_prefix = g_strdup_printf("%s ", cmd);
+
+    if (!g_str_has_prefix(input, cmd_prefix)) {
+        return NULL;
+    }
+
+    /* Extract provider name from input (after cmd_prefix) */
+    auto_gchar gchar* rest = g_strdup(input + strlen(cmd_prefix));
+    char* space = strchr(rest, ' ');
+    if (space) {
+        *space = '\0';
+    }
+
+    /* Look up the provider by name and get its models */
+    AIProvider* prov = ai_get_provider(rest);
+    if (!prov || !prov->models) {
+        return NULL;
+    }
+
+    if (!ai_models_ac) {
+        ai_models_ac = autocomplete_new();
+    }
+    /* Convert GList* to char** for autocomplete_update */
+    int model_count = g_list_length(prov->models);
+    char** model_array = g_new0(char*, model_count + 1);
+    GList* curr = prov->models;
+    int i = 0;
+    while (curr) {
+        model_array[i++] = curr->data;
+        curr = g_list_next(curr);
+    }
+
+    autocomplete_update(ai_models_ac, model_array);
+    g_free(model_array);
+
+    auto_gchar gchar* full_prefix = g_strdup_printf("%s%s", cmd_prefix, rest);
+
+    result = autocomplete_param_with_ac(input, full_prefix, ai_models_ac, TRUE, previous);
+    return result;
+}
--- a/src/command/cmd_defs.c
+++ b/src/command/cmd_defs.c
@@ -1289,8 +1289,8 @@ static const struct cmd_t command_defs[] = {
      CMD_TAGS(
              CMD_TAG_UI)
      CMD_SYN(
-              "/statusbar show name|number|read",
-              "/statusbar hide name|number|read",
+              "/statusbar show name|number|read|dbbackend",
+              "/statusbar hide name|number|read|dbbackend",
              "/statusbar maxtabs <value>",
              "/statusbar tablen <value>",
              "/statusbar tabmode default|dynamic|actlist",
@@ -1308,6 +1308,7 @@ static const struct cmd_t command_defs[] = {
              { "show|hide name", "Show or hide names in tabs." },
              { "show|hide number", "Show or hide numbers in tabs." },
              { "show|hide read", "Show or hide inactive tabs." },
+              { "show|hide dbbackend", "Show or hide the database backend indicator [sqlite] or [flatfile]." },
              { "self user|barejid|fulljid", "Show account user name, barejid, fulljid as status bar title." },
              { "self off", "Disable showing self as status bar title." },
              { "chat user|jid", "Show users name, or fulljid. Change needs a redraw/restart to take effect." },
@@ -1659,23 +1660,14 @@ static const struct cmd_t command_defs[] = {
    },

    { CMD_PREAMBLE("/logging",
-                   parse_args, 2, 3, &cons_logging_setting)
+                   parse_args, 0, 0, NULL)
      CMD_MAINFUNC(cmd_logging)
      CMD_TAGS(
              CMD_TAG_CHAT)
      CMD_SYN(
-              "/logging chat|group on|off")
+              "/logging")
      CMD_DESC(
-              "Configure chat logging. "
-              "Switch logging on or off. "
-              "Chat logging will be enabled if /history is set to on. "
-              "When disabling this option, /history will also be disabled. ")
-      CMD_ARGS(
-              { "chat on|off", "Enable/Disable regular chat logging." },
-              { "group on|off", "Enable/Disable groupchat (room) logging." })
-      CMD_EXAMPLES(
-              "/logging chat on",
-              "/logging group off")
+              "Deprecated. Use '/history' instead.")
    },

    { CMD_PREAMBLE("/states",
@@ -1888,15 +1880,16 @@ static const struct cmd_t command_defs[] = {
              "/history export [<jid>]",
              "/history import [<jid>]")
      CMD_DESC(
-              "Switch chat history on or off, /logging chat will automatically be enabled when this setting is on. "
-              "When history is enabled, previous messages are shown in chat windows. "
+              "Enable or disable chat history. "
+              "When on, messages are persisted via the active database backend and shown in chat windows. "
+              "When off, persistence stops and old history is hidden on chat-window open. "
              "Use 'backend' to show the active database backend. "
              "Use 'switch' to change the active database backend at runtime without reconnecting. "
              "Use 'verify' to check integrity of stored message history. "
              "Use 'export' to copy messages from SQLite to flat-file format, or 'import' to copy from flat-file to SQLite. "
              "Both export and import merge with existing data (duplicates are skipped).")
      CMD_ARGS(
-              { "on|off", "Enable or disable showing chat history." },
+              { "on|off", "Enable or disable persistence and showing chat history." },
              { "backend", "Show the name of the active database backend." },
              { "switch sqlite|flatfile", "Switch the active database backend at runtime." },
              { "verify [<jid>]", "Verify integrity of message history. Optionally specify a JID to check only one contact." },
@@ -2741,7 +2734,7 @@ static const struct cmd_t command_defs[] = {
              "clientid to set the client identification name "
              "session_alarm to configure an alarm when more clients log in.")
      CMD_ARGS(
-              { "logging on|redact|off|flatfile", "Switch chat logging. 'on' uses SQLite database (default). 'flatfile' stores messages as plain text files in ~/.local/share/profanity/flatlog/ that can be manually edited with any text editor. 'off' disables logging entirely. 'redact' stores messages with content replaced by '[redacted]'. Note: 'off' might have unintended consequences, such as not being able to decrypt OMEMO encrypted messages received later via MAM. The 'flatfile' setting takes effect on next connection." },
+              { "logging on|redact|off|flatfile", "Switch chat logging backend. 'on' uses the SQLite database (default). 'flatfile' stores messages as plain-text files under ~/.local/share/profanity/flatlog/ that can be edited with any text editor. 'off' disables logging entirely. 'redact' keeps logging but replaces every message body with '[REDACTED]'. Note: 'off' may have side effects such as not being able to decrypt OMEMO messages received later via MAM. Backend changes ('on' / 'flatfile') take effect immediately when connected, otherwise on next connect." },
              { "os on|off", "Choose whether to include the OS name if a user asks for software information (XEP-0092)." }
              )
      CMD_EXAMPLES(
@@ -2787,6 +2780,67 @@ static const struct cmd_t command_defs[] = {
              "/force-encryption policy block")
    },

+    { CMD_PREAMBLE("/ai",
+                   parse_args, 0, 5, NULL)
+      CMD_SUBFUNCS(
+              { "set", cmd_ai_set },
+              { "remove", cmd_ai_remove },
+              { "start", cmd_ai_start },
+              { "clear", cmd_ai_clear },
+              { "providers", cmd_ai_providers },
+              { "switch", cmd_ai_switch },
+              { "models", cmd_ai_models })
+      CMD_MAINFUNC(cmd_ai)
+      CMD_TAGS(
+              CMD_TAG_CHAT)
+      CMD_SYN(
+              "/ai",
+              "/ai set provider <name> <url>",
+              "/ai set token <provider> <token>",
+              "/ai set default-model <provider> <model>",
+              "/ai set custom <provider> <setting> <value>",
+              "/ai remove provider <name>",
+              "/ai providers",
+              "/ai start [<provider>] [<model>]",
+              "/ai switch <provider> [<model>]",
+              "/ai switch <model>",
+              "/ai models <provider>",
+              "/ai clear")
+      CMD_DESC(
+              "Interact with AI models via OpenAI-compatible APIs. "
+              "Supports multiple providers (openai, perplexity, custom). "
+              "Each provider has its own API key, endpoint, default model, and settings. "
+              "Chat history is maintained per session and not persisted locally.")
+      CMD_ARGS(
+              { "", "Display current AI settings and configured providers" },
+              { "set provider <name> <url>", "Add or update a provider with custom API endpoint" },
+              { "set token <provider> <token>", "Set API token for a provider (e.g., openai, perplexity)" },
+              { "set default-model <provider> <model>", "Set default model for a provider" },
+              { "set custom <provider> <setting> <value>", "Set provider-level setting (e.g., tools, search)" },
+              { "remove provider <name>", "Remove a custom provider" },
+              { "providers", "List configured providers with full details" },
+              { "start <provider> [<model>]", "Start new AI chat (space-separated, uses defaults if omitted)" },
+              { "switch <provider> [<model>]", "Switch to different provider (and optionally model)" },
+              { "models <provider>", "Fetch and display available models for provider" },
+              { "clear", "Clear current chat history" })
+      CMD_EXAMPLES(
+              "/ai",
+              "/ai set token openai sk-xxx",
+              "/ai set token perplexity pplx-xxx",
+              "/ai set provider custom https://my-api.com/v1/chat/completions",
+              "/ai set default-model perplexity sonar",
+              "/ai set custom perplexity tools enabled",
+              "/ai remove provider custom",
+              "/ai start",
+              "/ai start perplexity",
+              "/ai start perplexity sonar",
+              "/ai start openai gpt-4o",
+              "/ai switch gpt-4o",
+              "/ai switch openai gpt-5.4-nano",
+              "/ai models perplexity",
+              "/ai clear")
+    },
+
    // NEXT-COMMAND (search helper)
 };

--- a/src/command/cmd_funcs.c
+++ b/src/command/cmd_funcs.c
@@ -81,7 +81,6 @@
 #include "tools/bookmark_ignore.h"
 #include "tools/editor.h"
 #include "plugins/plugins.h"
-#include "ui/inputwin.h"
 #include "ui/ui.h"
 #include "ui/window_list.h"
 #include "xmpp/avatar.h"
@@ -117,6 +116,8 @@
 #include "tools/clipboard.h"
 #endif

+#include "ai/ai_client.h"
+
 #ifdef HAVE_PYTHON
 #include "plugins/python_plugins.h"
 #endif
@@ -6001,6 +6002,12 @@ cmd_statusbar(ProfWin* window, const char* const command, gchar** args)
            ui_resize();
            return TRUE;
        }
+        if (g_strcmp0(args[1], "dbbackend") == 0) {
+            prefs_set_boolean(PREF_STATUSBAR_SHOW_DBBACKEND, TRUE);
+            cons_show("Enabled showing database backend indicator.");
+            ui_resize();
+            return TRUE;
+        }
        cons_bad_cmd_usage(command);
        return TRUE;
    }
@@ -6034,6 +6041,12 @@ cmd_statusbar(ProfWin* window, const char* const command, gchar** args)
            ui_resize();
            return TRUE;
        }
+        if (g_strcmp0(args[1], "dbbackend") == 0) {
+            prefs_set_boolean(PREF_STATUSBAR_SHOW_DBBACKEND, FALSE);
+            cons_show("Disabled showing database backend indicator.");
+            ui_resize();
+            return TRUE;
+        }
        cons_bad_cmd_usage(command);
        return TRUE;
    }
@@ -6667,36 +6680,48 @@ cmd_privacy(ProfWin* window, const char* const command, gchar** args)
    }

    if (g_strcmp0(args[0], "logging") == 0) {
-        gchar* arg = args[1];
+        const gchar* arg = args[1];
        if (arg == NULL) {
            cons_bad_cmd_usage(command);
            return TRUE;
        }

-        if (g_strcmp0(arg, "on") == 0) {
-            cons_show("Logging enabled.");
-            prefs_set_string(PREF_DBLOG, arg);
-            prefs_set_boolean(PREF_CHLOG, TRUE);
-            prefs_set_boolean(PREF_HISTORY, TRUE);
-        } else if (g_strcmp0(arg, "off") == 0) {
-            cons_show("Logging disabled.");
-            prefs_set_string(PREF_DBLOG, arg);
-            prefs_set_boolean(PREF_CHLOG, FALSE);
-            prefs_set_boolean(PREF_HISTORY, FALSE);
-        } else if (g_strcmp0(arg, "redact") == 0) {
-            cons_show("Messages are going to be redacted.");
-            prefs_set_string(PREF_DBLOG, arg);
-        } else if (g_strcmp0(arg, "flatfile") == 0) {
-            auto_gchar gchar* ff_path = files_get_data_path(DIR_FLATLOG);
-            cons_show("Using flat-file backend for message logging. Takes effect on next connection.");
-            cons_show("Flatfile directory: %s", ff_path);
-            prefs_set_string(PREF_DBLOG, arg);
-            prefs_set_boolean(PREF_CHLOG, TRUE);
-            prefs_set_boolean(PREF_HISTORY, TRUE);
-        } else {
+        const gboolean is_on = (g_strcmp0(arg, "on") == 0);
+        const gboolean is_off = (g_strcmp0(arg, "off") == 0);
+        const gboolean is_redact = (g_strcmp0(arg, "redact") == 0);
+        const gboolean is_flatfile = (g_strcmp0(arg, "flatfile") == 0);
+
+        if (!is_on && !is_off && !is_redact && !is_flatfile) {
            cons_bad_cmd_usage(command);
            return TRUE;
        }
+
+        prefs_set_string(PREF_DBLOG, arg);
+        prefs_set_boolean(PREF_CHLOG, !is_off);
+        prefs_set_boolean(PREF_HISTORY, !is_off);
+
+        // For backend-switching values (on = sqlite, flatfile), apply the
+        // change immediately when connected so the user does not have to
+        // reconnect to see the new backend take effect. off/redact change
+        // write semantics but keep the currently-open backend alive.
+        if ((is_on || is_flatfile)
+            && active_db_backend
+            && connection_get_status() == JABBER_CONNECTED
+            && g_strcmp0(active_db_backend->name, is_on ? "sqlite" : "flatfile") != 0) {
+            log_database_switch_backend(is_on ? "on" : "flatfile");
+        }
+
+        if (is_on) {
+            cons_show("Database logging enabled (sqlite).");
+        } else if (is_off) {
+            cons_show("Database logging disabled.");
+        } else if (is_redact) {
+            cons_show("Database logging set to redact (message bodies replaced with [REDACTED]).");
+        } else if (is_flatfile) {
+            auto_gchar gchar* ff_path = files_get_data_path(DIR_FLATLOG);
+            cons_show("Database logging set to flat-file backend.");
+            cons_show("Flatfile directory: %s", ff_path);
+        }
        return TRUE;
    }

@@ -6706,26 +6731,7 @@ cmd_privacy(ProfWin* window, const char* const command, gchar** args)
 gboolean
 cmd_logging(ProfWin* window, const char* const command, gchar** args)
 {
-    if (args[0] == NULL) {
-        cons_logging_setting();
-        return TRUE;
-    }
-
-    if (strcmp(args[0], "chat") == 0 && args[1] != NULL) {
-        _cmd_set_boolean_preference(args[1], "Chat logging", PREF_CHLOG);
-
-        // if set to off, disable history
-        if (strcmp(args[1], "off") == 0) {
-            prefs_set_boolean(PREF_HISTORY, FALSE);
-        }
-
-        return TRUE;
-    } else if (g_strcmp0(args[0], "group") == 0 && args[1] != NULL) {
-        _cmd_set_boolean_preference(args[1], "Groupchat logging", PREF_GRLOG);
-        return TRUE;
-    }
-
-    cons_bad_cmd_usage(command);
+    cons_show("The '/logging' command is deprecated, use '/history' instead.");
    return TRUE;
 }

@@ -6872,9 +6878,20 @@ cmd_history(ProfWin* window, const char* const command, gchar** args)

    _cmd_set_boolean_preference(args[0], "Chat history", PREF_HISTORY);

-    // if set to on, set chlog (/logging chat on)
+    // /history off should stop persisting messages too, not just hide
+    // existing history on chat-window open. /history on resumes
+    // persistence (default backend) and parallel text chatlogs.
    if (strcmp(args[0], "on") == 0) {
        prefs_set_boolean(PREF_CHLOG, TRUE);
+        auto_gchar gchar* dblog = prefs_get_string(PREF_DBLOG);
+        if (g_strcmp0(dblog, "off") == 0) {
+            prefs_set_string(PREF_DBLOG, "on");
+            cons_show("Database logging re-enabled (sqlite).");
+        }
+    } else if (strcmp(args[0], "off") == 0) {
+        prefs_set_boolean(PREF_CHLOG, FALSE);
+        prefs_set_string(PREF_DBLOG, "off");
+        cons_show("Database logging disabled.");
    }

    return TRUE;
@@ -7291,7 +7308,7 @@ cmd_pgp(ProfWin* window, const char* const command, gchar** args)
            prefs_set_string(PREF_PGP_LOG, "on");
            cons_show("PGP messages will be logged as plaintext.");
            if (!prefs_get_boolean(PREF_CHLOG)) {
-                cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+                cons_show("Chat logging is currently disabled, use '/history on' to enable.");
            }
        } else if (g_strcmp0(choice, "off") == 0) {
            prefs_set_string(PREF_PGP_LOG, "off");
@@ -7300,7 +7317,7 @@ cmd_pgp(ProfWin* window, const char* const command, gchar** args)
            prefs_set_string(PREF_PGP_LOG, "redact");
            cons_show("PGP messages will be logged as '[redacted]'.");
            if (!prefs_get_boolean(PREF_CHLOG)) {
-                cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+                cons_show("Chat logging is currently disabled, use '/history on' to enable.");
            }
        } else {
            cons_bad_cmd_usage(command);
@@ -7787,7 +7804,7 @@ cmd_ox_log(ProfWin* window, const char* const command, gchar** args)
        prefs_set_string(PREF_OX_LOG, "on");
        cons_show("OX messages will be logged as plaintext.");
        if (!prefs_get_boolean(PREF_CHLOG)) {
-            cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+            cons_show("Chat logging is currently disabled, use '/history on' to enable.");
        }
    } else if (g_strcmp0(choice, "off") == 0) {
        prefs_set_string(PREF_OX_LOG, "off");
@@ -7796,7 +7813,7 @@ cmd_ox_log(ProfWin* window, const char* const command, gchar** args)
        prefs_set_string(PREF_OX_LOG, "redact");
        cons_show("OX messages will be logged as '[redacted]'.");
        if (!prefs_get_boolean(PREF_CHLOG)) {
-            cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+            cons_show("Chat logging is currently disabled, use '/history on' to enable.");
        }
    } else {
        cons_bad_cmd_usage(command);
@@ -7836,7 +7853,7 @@ cmd_otr_log(ProfWin* window, const char* const command, gchar** args)
        prefs_set_string(PREF_OTR_LOG, "on");
        cons_show("OTR messages will be logged as plaintext.");
        if (!prefs_get_boolean(PREF_CHLOG)) {
-            cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+            cons_show("Chat logging is currently disabled, use '/history on' to enable.");
        }
    } else if (g_strcmp0(choice, "off") == 0) {
        prefs_set_string(PREF_OTR_LOG, "off");
@@ -7845,7 +7862,7 @@ cmd_otr_log(ProfWin* window, const char* const command, gchar** args)
        prefs_set_string(PREF_OTR_LOG, "redact");
        cons_show("OTR messages will be logged as '[redacted]'.");
        if (!prefs_get_boolean(PREF_CHLOG)) {
-            cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+            cons_show("Chat logging is currently disabled, use '/history on' to enable.");
        }
    } else {
        cons_bad_cmd_usage(command);
@@ -8466,7 +8483,7 @@ _cmd_execute_default(ProfWin* window, const char* inp)
    }

    // handle non commands in non chat or plugin windows
-    if (window->type != WIN_CHAT && window->type != WIN_MUC && window->type != WIN_PRIVATE && window->type != WIN_PLUGIN && window->type != WIN_XML) {
+    if (window->type != WIN_CHAT && window->type != WIN_MUC && window->type != WIN_PRIVATE && window->type != WIN_PLUGIN && window->type != WIN_XML && window->type != WIN_AI) {
        cons_show("Unknown command: %s", inp);
        cons_alert(NULL);
        return TRUE;
@@ -8512,6 +8529,13 @@ _cmd_execute_default(ProfWin* window, const char* inp)
        connection_send_stanza(inp);
        break;
    }
+    case WIN_AI:
+    {
+        ProfAiWin* aiwin = (ProfAiWin*)window;
+        auto_gchar gchar* stanza_id = connection_create_stanza_id();
+        cl_ev_send_ai_msg(aiwin, inp, stanza_id);
+        break;
+    }
    default:
        break;
    }
@@ -8775,7 +8799,7 @@ cmd_omemo_log(ProfWin* window, const char* const command, gchar** args)
        prefs_set_string(PREF_OMEMO_LOG, "on");
        cons_show("OMEMO messages will be logged as plaintext.");
        if (!prefs_get_boolean(PREF_CHLOG)) {
-            cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+            cons_show("Chat logging is currently disabled, use '/history on' to enable.");
        }
    } else if (g_strcmp0(choice, "off") == 0) {
        prefs_set_string(PREF_OMEMO_LOG, "off");
@@ -8784,7 +8808,7 @@ cmd_omemo_log(ProfWin* window, const char* const command, gchar** args)
        prefs_set_string(PREF_OMEMO_LOG, "redact");
        cons_show("OMEMO messages will be logged as '[redacted]'.");
        if (!prefs_get_boolean(PREF_CHLOG)) {
-            cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+            cons_show("Chat logging is currently disabled, use '/history on' to enable.");
        }
    } else {
        cons_bad_cmd_usage(command);
@@ -10780,6 +10804,412 @@ cmd_vcard_save(ProfWin* window, const char* const command, gchar** args)
    return TRUE;
 }

+gboolean
+cmd_ai(ProfWin* window, const char* const command, gchar** args)
+{
+    if (args[0] == NULL) {
+        // Display current AI settings
+        cons_show("AI Chat - OpenAI-compatible API client");
+        cons_show("");
+
+        // List configured providers with full details
+        GList* providers = ai_list_providers();
+        cons_show("Configured providers:");
+        for (GList* curr = providers; curr; curr = g_list_next(curr)) {
+            AIProvider* provider = curr->data;
+            auto_gchar gchar* key = ai_get_provider_key(provider->name);
+            const gchar* default_model = ai_get_provider_default_model(provider->name);
+            cons_show("  %s", provider->name);
+            cons_show("    URL: %s", provider->api_url);
+            cons_show("    Key: %s", key ? "configured" : "NOT configured");
+            if (default_model) {
+                cons_show("    Default model: %s", default_model);
+            }
+            if (provider->models && g_list_length(provider->models) > 0) {
+                cons_show("    Cached models: %d", g_list_length(provider->models));
+            }
+            cons_show("");
+        }
+        g_list_free(providers);
+
+        cons_show("Use '/ai start' to begin a chat (uses default provider/model).");
+        cons_show("Use '/ai models <provider>' to fetch available models.");
+        return TRUE;
+    }
+
+    cons_bad_cmd_usage(command);
+    return TRUE;
+}
+
+gboolean
+cmd_ai_set(ProfWin* window, const char* const command, gchar** args)
+{
+    if (args[1] == NULL) {
+        cons_bad_cmd_usage(command);
+        return TRUE;
+    }
+
+    if (g_strcmp0(args[1], "provider") == 0) {
+        // /ai set provider <name> <url>
+        if (g_strv_length(args) < 4) {
+            cons_bad_cmd_usage(command);
+            return TRUE;
+        }
+        if (ai_add_provider(args[2], args[3])) {
+            cons_show("Provider '%s' configured with URL: %s", args[2], args[3]);
+        } else {
+            cons_show_error("Failed to configure provider '%s'.", args[2]);
+        }
+        cons_show("");
+        return TRUE;
+    } else if (g_strcmp0(args[1], "token") == 0) {
+        // /ai set token <provider> <token>
+        if (g_strv_length(args) < 4) {
+            cons_bad_cmd_usage(command);
+            return TRUE;
+        }
+        ai_set_provider_key(args[2], args[3]);
+        cons_show("API token set for provider: %s", args[2]);
+        cons_show("");
+        return TRUE;
+    } else if (g_strcmp0(args[1], "default-model") == 0) {
+        // /ai set default-model <provider> <model>
+        if (g_strv_length(args) < 4) {
+            cons_bad_cmd_usage(command);
+            return TRUE;
+        }
+        ai_set_provider_default_model(args[2], args[3]);
+        cons_show("Default model for provider '%s' set to: %s", args[2], args[3]);
+        cons_show("");
+        return TRUE;
+    } else if (g_strcmp0(args[1], "custom") == 0) {
+        // /ai set custom <provider> <setting> <value>
+        if (g_strv_length(args) < 5) {
+            cons_bad_cmd_usage(command);
+            return TRUE;
+        }
+        ai_set_provider_setting(args[2], args[3], args[4]);
+        cons_show("Setting '%s' for provider '%s' set to: %s", args[3], args[2], args[4]);
+        cons_show("");
+        return TRUE;
+    }
+
+    cons_bad_cmd_usage(command);
+    return TRUE;
+}
+
+gboolean
+cmd_ai_remove(ProfWin* window, const char* const command, gchar** args)
+{
+    // /ai remove provider <name>
+    if (g_strcmp0(args[1], "provider") != 0) {
+        cons_bad_cmd_usage(command);
+        return TRUE;
+    }
+
+    if (g_strv_length(args) < 3) {
+        cons_bad_cmd_usage(command);
+        return TRUE;
+    }
+
+    if (ai_remove_provider(args[2])) {
+        cons_show("Provider '%s' removed.", args[2]);
+    } else {
+        cons_show("Provider '%s' not found.", args[2]);
+    }
+    cons_show("");
+    return TRUE;
+}
+
+gboolean
+cmd_ai_start(ProfWin* window, const char* const command, gchar** args)
+{
+    // /ai start [<provider>] [<model>]
+    // Space-separated, no slash syntax
+    const gchar* provider_name = NULL;
+    const gchar* model = NULL;
+
+    if (g_strv_length(args) >= 2) {
+        provider_name = args[1];
+    }
+
+    if (!provider_name) {
+        cons_show_error("Please specify provider name.");
+        return TRUE;
+    }
+
+    if (g_strv_length(args) >= 3) {
+        model = args[2];
+    }
+
+    AIProvider* provider = ai_get_provider(provider_name);
+    if (!provider) {
+        cons_show_error("Provider '%s' not found. Use '/ai set provider %s <url>' to add it.",
+                        provider_name, provider_name);
+        return TRUE;
+    }
+
+    // Get model: explicit > provider default > hardcoded fallback
+    if (!model) {
+        model = ai_get_provider_default_model(provider_name);
+    }
+
+    if (!model) {
+        cons_show_error("No model specified and no default model set for provider '%s'.", provider_name);
+        cons_show("Use '/ai set default-model %s <model>'.", provider_name);
+        return TRUE;
+    }
+
+    // Check for API key
+    gchar* api_key = ai_get_provider_key(provider_name);
+    if (!api_key || strlen(api_key) == 0) {
+        cons_show_error("No API key set for provider '%s'. Use '/ai set token %s <key>' first.",
+                        provider_name, provider_name);
+        g_free(api_key);
+        return TRUE;
+    }
+    g_free(api_key);
+
+    // Create new AI session
+    AISession* session = ai_session_create(provider_name, model);
+    if (!session) {
+        log_error("[AI-CMD] Failed to create AI session");
+        cons_show_error("Failed to create AI session for %s/%s.", provider_name, model);
+        return TRUE;
+    }
+    log_debug("[AI-CMD] AI session created successfully");
+
+    // Create AI chat window
+    log_debug("[AI-CMD] Calling wins_new_ai()...");
+    ProfWin* ai_win = wins_new_ai(session);
+    if (!ai_win) {
+        log_error("[AI-CMD] wins_new_ai() returned NULL");
+        cons_show_error("Failed to create AI chat window.");
+        ai_session_unref(session);
+        return TRUE;
+    }
+    log_debug("[AI-CMD] wins_new_ai() returned successfully, window type: %d", ai_win->type);
+
+    // Release the reference held by cmd_ai_start, since the window now holds one
+    ai_session_unref(session);
+
+    // Add welcome message to the AI window
+    log_debug("[AI-CMD] Adding welcome messages...");
+    win_println(ai_win, THEME_DEFAULT, "-", "AI Chat: %s/%s", provider_name, model);
+    win_println(ai_win, THEME_DEFAULT, "-", "Type your message and press Enter to start a conversation.");
+    log_debug("[AI-CMD] Welcome messages added");
+
+    // Focus the new window
+    log_debug("[AI-CMD] Calling ui_focus_win()...");
+    ui_focus_win(ai_win);
+    log_debug("[AI-CMD] ui_focus_win() returned");
+
+    cons_show("Started AI chat: %s/%s", provider_name, model);
+    cons_show("");
+    log_debug("[AI-CMD] AI start command completed");
+
+    return TRUE;
+}
+
+gboolean
+cmd_ai_model(ProfWin* window, const char* const command, gchar** args)
+{
+    // /ai model <model>
+    if (args[1] == NULL) {
+        cons_bad_cmd_usage(command);
+        return TRUE;
+    }
+
+    const gchar* model = args[1];
+
+    // Must be in an AI window to use this command
+    if (!window || window->type != WIN_AI) {
+        cons_show_error("Must be in an AI chat window to use this command.");
+        return TRUE;
+    }
+    ProfAiWin* aiwin = (ProfAiWin*)window;
+
+    assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+
+    if (!aiwin->session) {
+        cons_show("No active session in this chat window.");
+        return TRUE;
+    }
+
+    ai_session_set_model(aiwin->session, model);
+    cons_show("Session model changed to: %s", model);
+    cons_show("");
+    return TRUE;
+}
+
+gboolean
+cmd_ai_switch(ProfWin* window, const char* const command, gchar** args)
+{
+    // /ai switch <provider> [<model>] - Change provider and optionally model
+    // /ai switch <model> - Change model only (keeps current provider)
+    // Modifies the existing session's provider and model instead of recreating it
+    if (g_strv_length(args) < 2) {
+        cons_bad_cmd_usage(command);
+        return TRUE;
+    }
+
+    // Must be in an AI window to use this command
+    if (!window || window->type != WIN_AI) {
+        cons_show_error("Must be in an AI chat window to use this command.");
+        return TRUE;
+    }
+    ProfAiWin* aiwin = (ProfAiWin*)window;
+
+    assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+
+    if (!aiwin->session) {
+        cons_show("No active session in this chat window.");
+        return TRUE;
+    }
+
+    const gchar* arg1 = args[1];
+    const gchar* arg2 = (g_strv_length(args) >= 3) ? args[2] : NULL;
+
+    const gchar* provider_name;
+    const gchar* model;
+
+    // Check if arg1 is a known provider
+    AIProvider* provider = ai_get_provider(arg1);
+    if (provider) {
+        // arg1 is a provider name
+        provider_name = arg1;
+
+        // Get model: arg2 > provider default > error
+        if (arg2) {
+            model = arg2;
+        } else {
+            model = ai_get_provider_default_model(provider_name);
+        }
+        if (!model) {
+            cons_show_error("No model specified and no default model set for provider '%s'.", provider_name);
+            cons_show("Use '/ai set default-model %s <model>' or '/ai switch %s <model>'.", provider_name, provider_name);
+            return TRUE;
+        }
+    } else {
+        // arg1 is a model name, keep current provider
+        provider_name = aiwin->session->provider_name;
+        if (!provider_name) {
+            cons_show_error("Session has no provider name.");
+            return TRUE;
+        }
+        model = arg1;
+    }
+
+    // Check for API key
+    auto_gchar gchar* api_key = ai_get_provider_key(provider_name);
+    if (!api_key || strlen(api_key) == 0) {
+        cons_show_error("No API key set for provider '%s'. Use '/ai set token %s <key>' first.",
+                        provider_name, provider_name);
+        return TRUE;
+    }
+
+    // Atomically switch session provider, model, and API key (thread-safe)
+    ai_session_switch(aiwin->session, provider_name, model, g_steal_pointer(&api_key));
+
+    // Update window title
+    win_println((ProfWin*)aiwin, THEME_DEFAULT, "-", "AI Chat: %s/%s", provider_name, model);
+
+    cons_show("Switched to %s/%s", provider_name, model);
+    cons_show("");
+
+    return TRUE;
+}
+
+gboolean
+cmd_ai_models(ProfWin* window, const char* const command, gchar** args)
+{
+    // /ai models <provider> [--cached]
+    // Default: fetch fresh models from API
+    // --cached: display cached models (if available)
+    if (args[1] == NULL) {
+        cons_bad_cmd_usage(command);
+        return TRUE;
+    }
+
+    const gchar* provider_name = args[1];
+    gboolean use_cached = (g_strv_length(args) >= 3) && (g_strcmp0(args[2], "--cached") == 0);
+
+    AIProvider* provider = ai_get_provider(provider_name);
+    if (!provider) {
+        cons_show_error("Provider '%s' not found.", provider_name);
+        return TRUE;
+    }
+
+    if (use_cached) {
+        // Display cached models (if available)
+        if (!provider->models) {
+            cons_show("No cached models for provider '%s'. Use '/ai models %s' to fetch from API.", provider_name, provider_name);
+            return TRUE;
+        }
+        cons_show("Cached models for provider '%s':", provider_name);
+        GList* curr = provider->models;
+        while (curr) {
+            cons_show("  %s", (gchar*)curr->data);
+            curr = g_list_next(curr);
+        }
+        cons_show("");
+        cons_show("Use '/ai models %s' to fetch fresh models from the API.", provider_name);
+    } else {
+        // Default: fetch fresh models from API
+        ai_fetch_models(provider_name, window);
+    }
+
+    return TRUE;
+}
+
+gboolean
+cmd_ai_clear(ProfWin* window, const char* const command, gchar** args)
+{
+    // /ai clear
+    // Must be in an AI window to use this command
+    if (!window || window->type != WIN_AI) {
+        cons_show_error("Must be in an AI chat window to use this command.");
+        return TRUE;
+    }
+    ProfAiWin* aiwin = (ProfAiWin*)window;
+
+    assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+    if (aiwin->session) {
+        ai_session_clear_history(aiwin->session);
+    }
+
+    win_clear(window);
+    cons_show("AI Chat history cleared.");
+    return TRUE;
+}
+
+gboolean
+cmd_ai_providers(ProfWin* window, const char* const command, gchar** args)
+{
+    cons_show("Configured providers:");
+    cons_show("");
+
+    GList* providers = ai_list_providers();
+    if (!providers) {
+        cons_show("  (none configured)");
+        cons_show("");
+        cons_show("Add a provider with: /ai set provider <name> <url>");
+        return TRUE;
+    }
+
+    for (GList* curr = providers; curr; curr = g_list_next(curr)) {
+        AIProvider* provider = curr->data;
+        auto_gchar gchar* key = ai_get_provider_key(provider->name);
+        cons_show("  %s", provider->name);
+        cons_show("    URL: %s", provider->api_url);
+        cons_show("    Key: %s", key ? "configured" : "NOT configured");
+        cons_show("");
+    }
+    g_list_free(providers);
+
+    return TRUE;
+}
+
 gboolean
 cmd_force_encryption(ProfWin* window, const char* const command, gchar** args)
 {
--- a/src/command/cmd_funcs.h
+++ b/src/command/cmd_funcs.h
@@ -166,6 +166,14 @@ gboolean cmd_console(ProfWin* window, const char* const command, gchar** args);
 gboolean cmd_command_list(ProfWin* window, const char* const command, gchar** args);
 gboolean cmd_command_exec(ProfWin* window, const char* const command, gchar** args);
 gboolean cmd_change_password(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai_set(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai_remove(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai_start(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai_clear(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai_providers(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai_switch(ProfWin* window, const char* const command, gchar** args);
+gboolean cmd_ai_models(ProfWin* window, const char* const command, gchar** args);

 gboolean cmd_plugins(ProfWin* window, const char* const command, gchar** args);
 gboolean cmd_plugins_sourcepath(ProfWin* window, const char* const command, gchar** args);
--- a/src/config/preferences.c
+++ b/src/config/preferences.c
@@ -66,6 +66,9 @@
 #define PREF_GROUP_MUC           "muc"
 #define PREF_GROUP_PLUGINS       "plugins"
 #define PREF_GROUP_EXECUTABLES   "executables"
+#define PREF_GROUP_AI            "ai"
+#define PREF_GROUP_AI_PREFIX     "ai/"      /* per-provider subsection prefix: [ai/<provider>] */
+#define PREF_AI_SETTING_PREFIX   "setting." /* per-provider custom setting key prefix */

 #define INPBLOCK_DEFAULT 1000

@@ -242,6 +245,69 @@ _prefs_load(void)
        g_key_file_remove_key(prefs, PREF_GROUP_UI, "titlebar.muc.title.jid", NULL);
        g_key_file_remove_key(prefs, PREF_GROUP_UI, "titlebar.muc.title.name", NULL);
    }
+
+    /* Migrate legacy flat [ai] layout
+     *   openai_url=..., openai=<token>, openai_models=..., openai_default_model=...
+     * into per-provider subsections
+     *   [ai/openai] url=..., key=..., models=..., default_model=...
+     * Idempotent: once the flat keys are gone, this is a no-op. */
+    if (g_key_file_has_group(prefs, PREF_GROUP_AI)) {
+        gsize key_count = 0;
+        auto_gcharv gchar** all_keys = g_key_file_get_keys(prefs, PREF_GROUP_AI, &key_count, NULL);
+
+        /* Collect provider names from <name>_url keys first so we can migrate
+         * the whole bundle (url + token + models + default_model) atomically
+         * per provider, without misclassifying unrelated flat keys. */
+        GHashTable* to_migrate = g_hash_table_new_full(g_str_hash, g_str_equal, g_free, NULL);
+        for (gsize i = 0; i < key_count; i++) {
+            const gchar* k = all_keys[i];
+            if (g_str_has_suffix(k, "_url") && !g_str_has_suffix(k, "_models_url")) {
+                gsize klen = strlen(k);
+                if (klen > 4) {
+                    g_hash_table_add(to_migrate, g_strndup(k, klen - 4));
+                }
+            }
+        }
+
+        GList* provider_names = g_hash_table_get_keys(to_migrate);
+        for (GList* curr = provider_names; curr; curr = g_list_next(curr)) {
+            const gchar* name = (const gchar*)curr->data;
+            auto_gchar gchar* group = g_strconcat(PREF_GROUP_AI_PREFIX, name, NULL);
+
+            const struct
+            {
+                const gchar* flat_suffix;
+                const gchar* new_key;
+            } map[] = {
+                { "_url", "url" },
+                { "_models", "models" },
+                { "_default_model", "default_model" },
+                { NULL, NULL },
+            };
+            for (int j = 0; map[j].flat_suffix; j++) {
+                auto_gchar gchar* flat_key = g_strconcat(name, map[j].flat_suffix, NULL);
+                if (g_key_file_has_key(prefs, PREF_GROUP_AI, flat_key, NULL)) {
+                    auto_gchar gchar* val = g_key_file_get_string(prefs, PREF_GROUP_AI, flat_key, NULL);
+                    if (val) {
+                        g_key_file_set_string(prefs, group, map[j].new_key, val);
+                    }
+                    g_key_file_remove_key(prefs, PREF_GROUP_AI, flat_key, NULL);
+                }
+            }
+
+            /* Token was stored as the bare provider name in [ai]. */
+            if (g_key_file_has_key(prefs, PREF_GROUP_AI, name, NULL)) {
+                auto_gchar gchar* val = g_key_file_get_string(prefs, PREF_GROUP_AI, name, NULL);
+                if (val) {
+                    g_key_file_set_string(prefs, group, "key", val);
+                }
+                g_key_file_remove_key(prefs, PREF_GROUP_AI, name, NULL);
+            }
+        }
+        g_list_free(provider_names);
+        g_hash_table_destroy(to_migrate);
+    }
+
    _save_prefs();

    boolean_choice_ac = autocomplete_new();
@@ -1706,6 +1772,366 @@ _save_prefs(void)
    save_keyfile(&prefs_prof_keyfile);
 }

+/* ========================================================================
+ * AI provider preferences
+ *
+ * Each provider lives in its own [ai/<name>] subsection of the keyfile with:
+ *   url=...
+ *   key=<api token>
+ *   default_model=...
+ *   models=model1;model2;...
+ *   setting.<custom>=value     (e.g. setting.tools=enabled)
+ * The bare [ai] section keeps only the global "defaults_initialized" sentinel.
+ * Legacy flat-key layouts are migrated on first load in _prefs_load().
+ * ======================================================================== */
+
+static gchar*
+_ai_section(const char* const provider)
+{
+    return g_strconcat(PREF_GROUP_AI_PREFIX, provider, NULL);
+}
+
+static gboolean
+_ai_set_string(const char* const provider, const char* const key, const char* const value)
+{
+    if (!provider || !key || !value || !prefs)
+        return FALSE;
+
+    auto_gchar gchar* group = _ai_section(provider);
+    g_key_file_set_string(prefs, group, key, value);
+    _save_prefs();
+    return TRUE;
+}
+
+static char*
+_ai_get_string(const char* const provider, const char* const key)
+{
+    if (!provider || !key || !prefs)
+        return NULL;
+
+    auto_gchar gchar* group = _ai_section(provider);
+    if (!g_key_file_has_key(prefs, group, key, NULL))
+        return NULL;
+    return g_key_file_get_string(prefs, group, key, NULL);
+}
+
+static gboolean
+_ai_remove_string(const char* const provider, const char* const key)
+{
+    if (!provider || !key || !prefs)
+        return FALSE;
+
+    auto_gchar gchar* group = _ai_section(provider);
+    if (!g_key_file_has_key(prefs, group, key, NULL))
+        return FALSE;
+
+    g_key_file_remove_key(prefs, group, key, NULL);
+    _save_prefs();
+    return TRUE;
+}
+
+/* --- Token --- */
+
+gboolean
+prefs_ai_set_token(const char* const provider, const char* const token)
+{
+    if (!provider || !prefs)
+        return FALSE;
+    if (!token)
+        return prefs_ai_remove_token(provider);
+    return _ai_set_string(provider, "key", token);
+}
+
+gboolean
+prefs_ai_remove_token(const char* const provider)
+{
+    return _ai_remove_string(provider, "key");
+}
+
+char*
+prefs_ai_get_token(const char* const provider)
+{
+    return _ai_get_string(provider, "key");
+}
+
+GList*
+prefs_ai_list_tokens(void)
+{
+    if (!prefs)
+        return NULL;
+
+    GList* result = NULL;
+    GList* providers = prefs_ai_list_providers();
+    for (GList* curr = providers; curr; curr = g_list_next(curr)) {
+        const gchar* name = (const gchar*)curr->data;
+        auto_gchar gchar* tok = prefs_ai_get_token(name);
+        if (tok && tok[0] != '\0') {
+            result = g_list_append(result, g_strdup(name));
+        }
+    }
+    prefs_free_ai_providers(providers);
+    return result;
+}
+
+void
+prefs_free_ai_tokens(GList* tokens)
+{
+    if (tokens) {
+        g_list_free_full(tokens, g_free);
+    }
+}
+
+/* --- Models cache --- */
+
+gboolean
+prefs_ai_set_models(const char* const provider, const gchar* const* models, gint count)
+{
+    if (!provider || count <= 0 || !prefs)
+        return FALSE;
+
+    GString* model_str = g_string_new("");
+    for (int i = 0; i < count; i++) {
+        if (i > 0)
+            g_string_append_c(model_str, ';');
+        g_string_append(model_str, models[i]);
+    }
+
+    gboolean ok = _ai_set_string(provider, "models", model_str->str);
+    g_string_free(model_str, TRUE);
+    return ok;
+}
+
+gboolean
+prefs_ai_remove_models(const char* const provider)
+{
+    return _ai_remove_string(provider, "models");
+}
+
+GList*
+prefs_ai_get_models(const char* const provider)
+{
+    auto_gchar gchar* model_str = _ai_get_string(provider, "models");
+    if (!model_str || model_str[0] == '\0')
+        return NULL;
+
+    GList* result = NULL;
+    gchar** models = g_strsplit(model_str, ";", -1);
+    for (int i = 0; models[i] != NULL; i++) {
+        result = g_list_append(result, g_strdup(models[i]));
+    }
+    g_strfreev(models);
+    return result;
+}
+
+void
+prefs_free_ai_models(GList* models)
+{
+    if (models) {
+        g_list_free_full(models, g_free);
+    }
+}
+
+/* --- Default model --- */
+
+gboolean
+prefs_ai_set_default_model(const char* const provider, const char* const model)
+{
+    if (!provider || !model || !prefs)
+        return FALSE;
+    if (model[0] == '\0')
+        return prefs_ai_remove_default_model(provider);
+    return _ai_set_string(provider, "default_model", model);
+}
+
+gboolean
+prefs_ai_remove_default_model(const char* const provider)
+{
+    return _ai_remove_string(provider, "default_model");
+}
+
+char*
+prefs_ai_get_default_model(const char* const provider)
+{
+    return _ai_get_string(provider, "default_model");
+}
+
+/* --- Provider URL --- */
+
+gboolean
+prefs_ai_set_provider(const char* const provider, const char* const url)
+{
+    if (!provider || !url)
+        return FALSE;
+    return _ai_set_string(provider, "url", url);
+}
+
+gboolean
+prefs_ai_remove_provider(const char* const provider)
+{
+    /* Provider presence is identified by the url key; removing it is what
+     * makes prefs_ai_list_providers stop returning this name. The remaining
+     * sub-keys are wiped through prefs_ai_remove_section(). */
+    return _ai_remove_string(provider, "url");
+}
+
+char*
+prefs_ai_get_provider_url(const char* const provider)
+{
+    return _ai_get_string(provider, "url");
+}
+
+gboolean
+prefs_ai_remove_section(const char* const provider)
+{
+    if (!provider || !prefs)
+        return FALSE;
+
+    auto_gchar gchar* group = _ai_section(provider);
+    if (!g_key_file_has_group(prefs, group))
+        return FALSE;
+
+    g_key_file_remove_group(prefs, group, NULL);
+    _save_prefs();
+    return TRUE;
+}
+
+GList*
+prefs_ai_list_providers(void)
+{
+    if (!prefs)
+        return NULL;
+
+    GList* result = NULL;
+    gsize len = 0;
+    auto_gcharv gchar** groups = g_key_file_get_groups(prefs, &len);
+    if (!groups)
+        return NULL;
+
+    gsize prefix_len = strlen(PREF_GROUP_AI_PREFIX);
+    for (gsize i = 0; i < len; i++) {
+        const gchar* g = groups[i];
+        if (g_str_has_prefix(g, PREF_GROUP_AI_PREFIX) && strlen(g) > prefix_len) {
+            /* Only count groups that actually identify a provider (have url). */
+            if (g_key_file_has_key(prefs, g, "url", NULL)) {
+                result = g_list_append(result, g_strdup(g + prefix_len));
+            }
+        }
+    }
+
+    return result;
+}
+
+void
+prefs_free_ai_providers(GList* providers)
+{
+    if (providers) {
+        g_list_free_full(providers, g_free);
+    }
+}
+
+GList*
+prefs_ai_get_providers(void)
+{
+    GList* configured = prefs_ai_list_providers();
+    GList* result = NULL;
+
+    for (GList* curr = configured; curr; curr = g_list_next(curr)) {
+        const gchar* name = (const gchar*)curr->data;
+        auto_gchar gchar* url = prefs_ai_get_provider_url(name);
+        if (url && strlen(url) > 0) {
+            result = g_list_append(result, g_strdup(name));
+            result = g_list_append(result, g_strdup(url));
+        }
+    }
+    prefs_free_ai_providers(configured);
+
+    /* Seed openai/perplexity into the in-memory result on very first init
+     * only. In production the "defaults_initialized" sentinel in [ai] is what
+     * tells us we've already seeded — a user who removes every provider must
+     * not get them silently re-added on the next start. When prefs is NULL
+     * (e.g. unit tests that bring up ai_client without prefs_load) we still
+     * seed the in-memory list so behaviour matches a fresh first run, but
+     * skip the persistence write. */
+    gboolean already_seeded = prefs && g_key_file_get_boolean(prefs, PREF_GROUP_AI, "defaults_initialized", NULL);
+    if (!already_seeded && !result) {
+        if (prefs) {
+            prefs_ai_set_provider("openai", "https://api.openai.com/");
+            prefs_ai_set_provider("perplexity", "https://api.perplexity.ai/");
+        }
+        result = g_list_append(result, g_strdup("openai"));
+        result = g_list_append(result, g_strdup("https://api.openai.com/"));
+        result = g_list_append(result, g_strdup("perplexity"));
+        result = g_list_append(result, g_strdup("https://api.perplexity.ai/"));
+    }
+    if (prefs && !already_seeded) {
+        g_key_file_set_boolean(prefs, PREF_GROUP_AI, "defaults_initialized", TRUE);
+        _save_prefs();
+    }
+
+    return result;
+}
+
+/* --- Custom per-provider settings (setting.<name>) --- */
+
+gboolean
+prefs_ai_set_setting(const char* const provider, const char* const setting, const char* const value)
+{
+    if (!provider || !setting)
+        return FALSE;
+    auto_gchar gchar* key = g_strconcat(PREF_AI_SETTING_PREFIX, setting, NULL);
+    if (!value)
+        return _ai_remove_string(provider, key);
+    return _ai_set_string(provider, key, value);
+}
+
+gboolean
+prefs_ai_remove_setting(const char* const provider, const char* const setting)
+{
+    if (!provider || !setting)
+        return FALSE;
+    auto_gchar gchar* key = g_strconcat(PREF_AI_SETTING_PREFIX, setting, NULL);
+    return _ai_remove_string(provider, key);
+}
+
+char*
+prefs_ai_get_setting(const char* const provider, const char* const setting)
+{
+    if (!provider || !setting)
+        return NULL;
+    auto_gchar gchar* key = g_strconcat(PREF_AI_SETTING_PREFIX, setting, NULL);
+    return _ai_get_string(provider, key);
+}
+
+GList*
+prefs_ai_list_settings(const char* const provider)
+{
+    if (!provider || !prefs)
+        return NULL;
+
+    auto_gchar gchar* group = _ai_section(provider);
+    if (!g_key_file_has_group(prefs, group))
+        return NULL;
+
+    GList* result = NULL;
+    gsize len = 0;
+    auto_gcharv gchar** keys = g_key_file_get_keys(prefs, group, &len, NULL);
+    gsize prefix_len = strlen(PREF_AI_SETTING_PREFIX);
+    for (gsize i = 0; i < len; i++) {
+        if (g_str_has_prefix(keys[i], PREF_AI_SETTING_PREFIX) && strlen(keys[i]) > prefix_len) {
+            result = g_list_append(result, g_strdup(keys[i] + prefix_len));
+        }
+    }
+    return result;
+}
+
+void
+prefs_free_ai_settings(GList* settings)
+{
+    if (settings) {
+        g_list_free_full(settings, g_free);
+    }
+}
+
 // get the preference group for a specific preference
 // for example the PREF_BEEP setting ("beep" in .profrc, see _get_key) belongs
 // to the [ui] section.
@@ -1782,6 +2208,7 @@ _get_group(preference_t pref)
    case PREF_STATUSBAR_SHOW_NAME:
    case PREF_STATUSBAR_SHOW_NUMBER:
    case PREF_STATUSBAR_SHOW_READ:
+    case PREF_STATUSBAR_SHOW_DBBACKEND:
    case PREF_STATUSBAR_SELF:
    case PREF_STATUSBAR_CHAT:
    case PREF_STATUSBAR_ROOM_TITLE:
@@ -1865,6 +2292,9 @@ _get_group(preference_t pref)
        return PREF_GROUP_OMEMO;
    case PREF_OX_LOG:
        return PREF_GROUP_OX;
+    case PREF_AI_PROVIDER:
+    case PREF_AI_API_KEY:
+        return PREF_GROUP_AI;
    default:
        return NULL;
    }
@@ -2106,6 +2536,8 @@ _get_key(preference_t pref)
        return "statusbar.show.number";
    case PREF_STATUSBAR_SHOW_READ:
        return "statusbar.show.read";
+    case PREF_STATUSBAR_SHOW_DBBACKEND:
+        return "statusbar.show.dbbackend";
    case PREF_STATUSBAR_SELF:
        return "statusbar.self";
    case PREF_STATUSBAR_CHAT:
@@ -2142,6 +2574,10 @@ _get_key(preference_t pref)
        return "stamp.incoming";
    case PREF_OX_LOG:
        return "log";
+    case PREF_AI_PROVIDER:
+        return "provider";
+    case PREF_AI_API_KEY:
+        return "api_key";
    case PREF_MOOD:
        return "mood";
    case PREF_VCARD_PHOTO_CMD:
@@ -2195,6 +2631,7 @@ _get_default_boolean(preference_t pref)
    case PREF_ROOM_LIST_CACHE:
    case PREF_STATUSBAR_SHOW_NUMBER:
    case PREF_STATUSBAR_SHOW_READ:
+    case PREF_STATUSBAR_SHOW_DBBACKEND:
    case PREF_REVEAL_OS:
    case PREF_CORRECTION_ALLOW:
    case PREF_RECEIPTS_SEND:
@@ -2319,6 +2756,10 @@ _get_default_string(preference_t pref)
        return "on";
    case PREF_FORCE_ENCRYPTION_MODE:
        return "resend-to-confirm";
+    case PREF_AI_PROVIDER:
+        return "openai";
+    case PREF_AI_API_KEY:
+        return "";
    default:
        return NULL;
    }
--- a/src/config/preferences.h
+++ b/src/config/preferences.h
@@ -162,6 +162,7 @@ typedef enum {
    PREF_STATUSBAR_SHOW_NAME,
    PREF_STATUSBAR_SHOW_NUMBER,
    PREF_STATUSBAR_SHOW_READ,
+    PREF_STATUSBAR_SHOW_DBBACKEND,
    PREF_STATUSBAR_SELF,
    PREF_STATUSBAR_CHAT,
    PREF_STATUSBAR_ROOM_TITLE,
@@ -183,6 +184,9 @@ typedef enum {
    PREF_OX_LOG,
    PREF_MOOD,
    PREF_STROPHE_VERBOSITY,
+    PREF_AI_PROVIDER,
+    PREF_AI_API_KEY,
+    PREF_AI_DEFAULT_MODEL,
    PREF_STROPHE_SM_ENABLED,
    PREF_STROPHE_SM_RESEND,
    PREF_VCARD_PHOTO_CMD,
@@ -356,4 +360,42 @@ gboolean prefs_get_room_notify(const char* const roomjid);
 gboolean prefs_get_room_notify_mention(const char* const roomjid);
 gboolean prefs_get_room_notify_trigger(const char* const roomjid);

+/* AI token management */
+gboolean prefs_ai_set_token(const char* const provider, const char* const token);
+gboolean prefs_ai_remove_token(const char* const provider);
+char* prefs_ai_get_token(const char* const provider);
+GList* prefs_ai_list_tokens(void);
+void prefs_free_ai_tokens(GList* tokens);
+
+/* AI provider management */
+gboolean prefs_ai_set_provider(const char* const provider, const char* const url);
+gboolean prefs_ai_remove_provider(const char* const provider);
+char* prefs_ai_get_provider_url(const char* const provider);
+GList* prefs_ai_list_providers(void);
+void prefs_free_ai_providers(GList* providers);
+GList* prefs_ai_get_providers(void);
+
+/* Wipe the whole [ai/<provider>] subsection (url, key, models, default_model,
+ * custom settings) in one call. */
+gboolean prefs_ai_remove_section(const char* const provider);
+
+/* AI model cache management */
+gboolean prefs_ai_set_models(const char* const provider, const gchar* const* models, gint count);
+gboolean prefs_ai_remove_models(const char* const provider);
+GList* prefs_ai_get_models(const char* const provider);
+void prefs_free_ai_models(GList* models);
+
+/* AI default model per provider */
+gboolean prefs_ai_set_default_model(const char* const provider, const char* const model);
+gboolean prefs_ai_remove_default_model(const char* const provider);
+char* prefs_ai_get_default_model(const char* const provider);
+
+/* AI per-provider custom settings (e.g. tools, search) — persisted as
+ * setting.<name>=<value> inside [ai/<provider>]. Pass NULL value to remove. */
+gboolean prefs_ai_set_setting(const char* const provider, const char* const setting, const char* const value);
+gboolean prefs_ai_remove_setting(const char* const provider, const char* const setting);
+char* prefs_ai_get_setting(const char* const provider, const char* const setting);
+GList* prefs_ai_list_settings(const char* const provider);
+void prefs_free_ai_settings(GList* settings);
+
 #endif
--- a/src/database.c
+++ b/src/database.c
@@ -61,6 +61,22 @@ integrity_issue_free(integrity_issue_t* issue)
    }
 }

+gboolean
+log_database_can_recover_messages(void)
+{
+    if (!active_db_backend) {
+        return FALSE;
+    }
+    auto_gchar gchar* pref_dblog = prefs_get_string(PREF_DBLOG);
+    if (g_strcmp0(pref_dblog, "off") == 0) {
+        return FALSE;
+    }
+    if (g_strcmp0(pref_dblog, "redact") == 0) {
+        return FALSE;
+    }
+    return TRUE;
+}
+
 gboolean
 log_database_init(ProfAccount* account)
 {
--- a/src/database.h
+++ b/src/database.h
@@ -109,6 +109,9 @@ ProfMessage* log_database_get_limits_info(const gchar* const contact_barejid, gb
 void log_database_close(void);
 GSList* log_database_verify_integrity(const gchar* const contact_barejid);

+// FALSE if no backend, or PREF_DBLOG is "off" / "redact".
+gboolean log_database_can_recover_messages(void);
+
 // Cross-backend export/import (requires HAVE_SQLITE)
 #ifdef HAVE_SQLITE
 int log_database_export_to_flatfile(const gchar* const contact_jid);
--- a/src/database_flatfile.c
+++ b/src/database_flatfile.c
@@ -88,9 +88,8 @@ _ff_cache_line_ids(ff_contact_state_t* state, const char* line)
        return;

    // Split metadata on unescaped '|'
-    char* meta_str = g_strndup(bracket + 1, close - bracket - 1);
+    auto_gchar gchar* meta_str = g_strndup(bracket + 1, close - bracket - 1);
    char** parts = ff_split_meta(meta_str);
-    g_free(meta_str);

    char* stanza_id = NULL;
    char* archive_id = NULL;
@@ -141,15 +140,13 @@ _ff_maybe_index_line(ff_contact_state_t* state, const char* buf, off_t pos)
    if (!space)
        return;

-    char* ts_str = g_strndup(buf, space - buf);
+    auto_gchar gchar* ts_str = g_strndup(buf, space - buf);
    GDateTime* dt = g_date_time_new_from_iso8601(ts_str, NULL);
    if (!dt) {
        log_warning("flatfile: unparsable timestamp in %s at offset %ld: %s",
                    state->filepath, (long)pos, ts_str);
-        g_free(ts_str);
        return;
    }
-    g_free(ts_str);

    if (state->n_entries >= state->cap_entries) {
        state->cap_entries = state->cap_entries ? state->cap_entries * 2 : 64;
--- a/src/event/client_events.c
+++ b/src/event/client_events.c
@@ -37,8 +37,10 @@
 #include "config.h"

 #include <stdlib.h>
+#include <assert.h>
 #include <glib.h>

+#include "ai/ai_client.h"
 #include "log.h"
 #include "chatlog.h"
 #include "database.h"
@@ -47,7 +49,7 @@
 #include "event/common.h"
 #include "plugins/plugins.h"
 #include "ui/inputwin.h"
-#include "ui/window_list.h"
+#include "ui/window.h"
 #include "xmpp/chat_session.h"
 #include "xmpp/session.h"
 #include "xmpp/xmpp.h"
@@ -284,3 +286,24 @@ allow_unencrypted_message(ProfChatWin* chatwin, const char* const msg)
    win_println((ProfWin*)chatwin, THEME_ERROR, "-", "Message not sent: invalid encryption mode (%s). Use '/force-encryption policy resend-to-confirm'.", force_enc_mode);
    return FALSE;
 }
+
+void
+cl_ev_send_ai_msg(ProfAiWin* aiwin, const char* const message, const char* const id)
+{
+    if (!aiwin || !message) {
+        return;
+    }
+
+    assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+
+    if (!aiwin->session) {
+        log_error("[AI] AI session not initialized");
+        return;
+    }
+
+    // Display user message in AI window.
+    win_print_outgoing(&aiwin->window, ">>", id, NULL, message);
+
+    // Send to AI provider.
+    ai_send_prompt(aiwin->session, message, aiwin);
+}
--- a/src/event/client_events.h
+++ b/src/event/client_events.h
@@ -51,6 +51,7 @@ void cl_ev_send_msg(ProfChatWin* chatwin, const char* const msg, const char* con
 void cl_ev_send_muc_msg_corrected(ProfMucWin* mucwin, const char* const msg, const char* const oob_url, gboolean correct_last_msg);
 void cl_ev_send_muc_msg(ProfMucWin* mucwin, const char* const msg, const char* const oob_url);
 void cl_ev_send_priv_msg(ProfPrivateWin* privwin, const char* const msg, const char* const oob_url);
+void cl_ev_send_ai_msg(ProfAiWin* aiwin, const char* const message, const char* const id);

 // Checks if an unencrypted message can be sent based on encryption preferences.
 // Returns TRUE if allowed, FALSE if blocked.
--- a/src/profanity.c
+++ b/src/profanity.c
@@ -69,6 +69,7 @@
 #include "xmpp/xmpp.h"
 #include "xmpp/muc.h"
 #include "xmpp/chat_session.h"
+#include "ai/ai_client.h"
 #include "xmpp/chat_state.h"
 #include "xmpp/contact.h"
 #include "xmpp/roster_list.h"
@@ -249,6 +250,7 @@ _init(char* log_level, char* config_file, char* log_file, char* theme_name)
    }
    session_init();
    cmd_init();
+    ai_client_init();
    log_info("Initialising contact list");
    muc_init();
    tlscerts_init();
--- a/src/tools/autocomplete.c
+++ b/src/tools/autocomplete.c
@@ -260,7 +260,7 @@ autocomplete_complete(Autocomplete ac, const gchar* search_str, gboolean quote,
            FREE_SET_NULL(ac->search_str);
        }

-        ac->search_str = strdup(search_str);
+        ac->search_str = g_strdup(search_str);
        found = _search(ac, ac->items, quote, NEXT);

        return found;
@@ -305,7 +305,7 @@ autocomplete_complete(Autocomplete ac, const gchar* search_str, gboolean quote,
 // autocomplete_func func is used -> autocomplete_param_with_func
 // Autocomplete ac, gboolean quote are used -> autocomplete_param_with_ac
 static char*
-_autocomplete_param_common(const char* const input, char* command, autocomplete_func func, Autocomplete ac, gboolean quote, gboolean previous, void* context)
+_autocomplete_param_common(const char* const input, const char* command, autocomplete_func func, Autocomplete ac, gboolean quote, gboolean previous, void* context)
 {
    int len;
    auto_gchar gchar* command_cpy = g_strdup_printf("%s ", command);
@@ -344,7 +344,7 @@ _autocomplete_param_common(const char* const input, char* command, autocomplete_
 }

 char*
-autocomplete_param_with_func(const char* const input, char* command, autocomplete_func func, gboolean previous, void* context)
+autocomplete_param_with_func(const char* const input, const char* command, autocomplete_func func, gboolean previous, void* context)
 {
    return _autocomplete_param_common(input, command, func, NULL, FALSE, previous, context);
 }
--- a/src/tools/autocomplete.h
+++ b/src/tools/autocomplete.h
@@ -63,7 +63,7 @@ gchar* autocomplete_complete(Autocomplete ac, const gchar* search_str, gboolean
 GList* autocomplete_create_list(Autocomplete ac);
 gint autocomplete_length(Autocomplete ac);

-char* autocomplete_param_with_func(const char* const input, char* command,
+char* autocomplete_param_with_func(const char* const input, const char* command,
                                   autocomplete_func func, gboolean previous, void* context);

 char* autocomplete_param_with_ac(const char* const input, char* command,
--- a/src/ui/aiwin.c
+++ b/src/ui/aiwin.c
@@ -0,0 +1,61 @@
+/*
+ * aiwin.c - AI Chat Window Management
+ *
+ * Provides functions for managing the AI chat window (ProfAiWin) in the
+ * profanity client. This includes retrieving window strings for identification,
+ * displaying AI/LLM responses in the chat view, and handling error messages
+ * from AI interactions.
+ *
+ * Key functions:
+ *   aiwin_get_string()       - Returns a descriptive string for the window
+ *   aiwin_display_response() - Displays an AI/LLM response and appends it
+ *                              to the conversation history
+ *   aiwin_display_error()    - Displays an error message from AI operations
+ *
+ * vim: expandtab:ts=4:sts=4:sw=4
+ */
+
+#include "config.h"
+
+#include <assert.h>
+
+#include "ai/ai_client.h"
+#include "log.h"
+#include "ui/ui.h"
+#include "ui/win_types.h"
+
+char*
+aiwin_get_string(ProfAiWin* win)
+{
+    assert(win->memcheck == PROFAIWIN_MEMCHECK);
+    return g_strdup_printf("AI Chat (%s: %s)", win->session->provider_name, win->session->model);
+}
+
+void
+aiwin_display_response(ProfAiWin* win, const char* response)
+{
+    log_debug("[AI-WIN] aiwin_display_response ENTER: win=%p, response='%s'", (void*)win, response);
+    assert(win->memcheck == PROFAIWIN_MEMCHECK);
+
+    if (!response) {
+        log_error("[AI-WIN] FAIL: null response");
+        return;
+    }
+
+    // Display AI response
+    win_println(&win->window, THEME_DEFAULT, "<< LLM:", "%s", response);
+    log_debug("[AI-WIN] Displayed AI response");
+}
+
+void
+aiwin_display_error(ProfAiWin* win, const char* error_msg)
+{
+    log_debug("[AI-WIN] aiwin_display_error ENTER: win=%p, error='%s'", (void*)win, error_msg);
+    assert(win->memcheck == PROFAIWIN_MEMCHECK);
+
+    const char* msg = error_msg ? error_msg : "Unknown error";
+    win_println(&win->window, THEME_ERROR, "[ERROR]", "%s", msg);
+
+    // Also show error in console so user sees it regardless of active window
+    cons_show_error("AI Error: %s", msg);
+}
--- a/src/ui/console.c
+++ b/src/ui/console.c
@@ -1884,6 +1884,12 @@ cons_statusbar_setting(void)
        cons_show("Show tab with no actions (/statusbar)       : OFF");
    }

+    if (prefs_get_boolean(PREF_STATUSBAR_SHOW_DBBACKEND)) {
+        cons_show("Show database backend (/statusbar)          : ON");
+    } else {
+        cons_show("Show database backend (/statusbar)          : OFF");
+    }
+
    cons_show("Max tabs (/statusbar)                       : %d", prefs_get_statusbartabs());

    gint pref_len = prefs_get_statusbartablen();
@@ -1942,27 +1948,12 @@ cons_log_setting(void)
    cons_show("Log level (/log level)      : %s", level);
 }

-void
-cons_logging_setting(void)
-{
-    if (prefs_get_boolean(PREF_CHLOG))
-        cons_show("Chat logging (/logging chat)                : ON");
-    else
-        cons_show("Chat logging (/logging chat)                : OFF");
-
-    if (prefs_get_boolean(PREF_GRLOG))
-        cons_show("Groupchat logging (/logging group)          : ON");
-    else
-        cons_show("Groupchat logging (/logging group)          : OFF");
-}
-
 void
 cons_show_log_prefs(void)
 {
    cons_show("Logging preferences:");
    cons_show("");
    cons_log_setting();
-    cons_logging_setting();

    cons_alert(NULL);
 }
@@ -2856,11 +2847,6 @@ cons_privacy_setting(void)
 {
    cons_show("Database logging                                           : %s", prefs_get_string(PREF_DBLOG));

-    if (prefs_get_boolean(PREF_CHLOG)) {
-        cons_show("Chat logging (/logging chat)                               : ON");
-    } else {
-        cons_show("Chat logging (/logging chat)                               : OFF");
-    }
    if (prefs_get_boolean(PREF_HISTORY)) {
        cons_show("Chat history (/history)                                    : ON");
    } else {
--- a/src/ui/statusbar.c
+++ b/src/ui/statusbar.c
@@ -609,6 +609,8 @@ _status_bar_draw_maintext(int pos)
 static int
 _status_bar_draw_dbbackend(int pos)
 {
+    if (!prefs_get_boolean(PREF_STATUSBAR_SHOW_DBBACKEND))
+        return pos;
    if (!active_db_backend || !active_db_backend->name)
        return pos;

--- a/src/ui/titlebar.c
+++ b/src/ui/titlebar.c
@@ -46,6 +46,7 @@
 #include "ui/ui.h"
 #include "ui/titlebar.h"
 #include "ui/inputwin.h"
+#include "ui/win_types.h"
 #include "ui/window_list.h"
 #include "ui/window.h"
 #include "ui/screen.h"
@@ -252,6 +253,10 @@ _title_bar_draw(void)
        _show_muc_privacy(mucwin);
        _show_attention(current, mucwin->has_attention);
        _show_scrolled(current);
+    } else if (current && current->type == WIN_AI) {
+        ProfAiWin* aiwin = (ProfAiWin*)current;
+        assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+        _show_scrolled(current);
    }

    _show_self_presence();
--- a/src/ui/ui.h
+++ b/src/ui/ui.h
@@ -330,7 +330,6 @@ void cons_history_setting(void);
 void cons_carbons_setting(void);
 void cons_receipts_setting(void);
 void cons_log_setting(void);
-void cons_logging_setting(void);
 void cons_autoaway_setting(void);
 void cons_reconnect_setting(void);
 void cons_autoping_setting(void);
@@ -386,6 +385,8 @@ ProfWin* win_create_config(const char* const title, DataForm* form, ProfConfWinC
 ProfWin* win_create_private(const char* const fulljid);
 ProfWin* win_create_plugin(const char* const plugin_name, const char* const tag);
 ProfWin* win_create_vcard(vCard* vcard);
+ProfWin* win_create_ai(AISession* session);
+ProfWin* wins_new_ai(AISession* session);
 void win_update_virtual(ProfWin* window);
 void win_free(ProfWin* window);
 gboolean win_notify_remind(ProfWin* window);
@@ -445,4 +446,9 @@ void notify_invite(const char* const from, const char* const room, const char* c
 void notify(const char* const message, int timeout, const char* const category);
 void notify_subscription(const char* const from);

+/* AI window */
+char* aiwin_get_string(ProfAiWin* win);
+void aiwin_display_response(ProfAiWin* win, const char* response);
+void aiwin_display_error(ProfAiWin* win, const char* error_msg);
+
 #endif
--- a/src/ui/win_types.h
+++ b/src/ui/win_types.h
@@ -62,6 +62,7 @@
 #define PROFXMLWIN_MEMCHECK     87333463
 #define PROFPLUGINWIN_MEMCHECK  43434777
 #define PROFVCARDWIN_MEMCHECK   68947523
+#define PROFAIWIN_MEMCHECK      91827364

 typedef enum {
    FIELD_HIDDEN,
@@ -144,7 +145,8 @@ typedef enum {
    WIN_PRIVATE,
    WIN_XML,
    WIN_PLUGIN,
-    WIN_VCARD
+    WIN_VCARD,
+    WIN_AI
 } win_type_t;

 typedef enum {
@@ -258,4 +260,16 @@ typedef struct prof_vcard_win_t
    unsigned long memcheck;
 } ProfVcardWin;

+/* Forward declaration */
+typedef struct ai_session_t AISession;
+
+typedef struct prof_ai_win_t
+{
+    ProfWin window;
+    ProfBuff message_buffer;
+    GString* conversation_history;
+    AISession* session;
+    unsigned long memcheck;
+} ProfAiWin;
+
 #endif
--- a/src/ui/window.c
+++ b/src/ui/window.c
@@ -36,6 +36,7 @@

 #include "config.h"
 #include "database.h"
+#include "ui/win_types.h"
 #include "ui/window_list.h"

 #include <stdlib.h>
@@ -62,6 +63,7 @@
 #include "ui/screen.h"
 #include "xmpp/xmpp.h"
 #include "xmpp/roster_list.h"
+#include "ai/ai_client.h"

 static const char* LOADING_MESSAGE = "Loading older messages…";
 static const char* CONS_WIN_TITLE = "CProof. Type /help for help information.";
@@ -329,6 +331,22 @@ win_create_vcard(vCard* vcard)
    return &new_win->window;
 }

+ProfWin*
+win_create_ai(AISession* session)
+{
+    ProfAiWin* new_win = g_new0(ProfAiWin, 1);
+    new_win->window.type = WIN_AI;
+    new_win->window.scroll_state = WIN_SCROLL_INNER;
+    new_win->window.layout = _win_create_simple_layout();
+
+    new_win->message_buffer = buffer_create();
+    new_win->conversation_history = g_string_new("");
+    new_win->session = session ? ai_session_ref(session) : NULL;
+    new_win->memcheck = PROFAIWIN_MEMCHECK;
+
+    return &new_win->window;
+}
+
 gchar*
 win_get_title(ProfWin* window)
 {
@@ -407,6 +425,13 @@ win_get_title(ProfWin* window)

        return g_string_free(title, FALSE);
    }
+    case WIN_AI:
+    {
+        const ProfAiWin* aiwin = (ProfAiWin*)window;
+        assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+
+        return g_strdup_printf("AI Assistant (%s: %s)", aiwin->session->provider_name, aiwin->session->model);
+    }
    }
    assert(FALSE);
 }
@@ -454,6 +479,10 @@ win_get_tab_identifier(ProfWin* window)
    {
        return strdup("vcard");
    }
+    case WIN_AI:
+    {
+        return strdup("ai");
+    }
    }
    assert(FALSE);
 }
@@ -536,6 +565,12 @@ win_to_string(ProfWin* window)
        ProfVcardWin* vcardwin = (ProfVcardWin*)window;
        return vcardwin_get_string(vcardwin);
    }
+    case WIN_AI:
+    {
+        ProfAiWin* aiwin = (ProfAiWin*)window;
+        assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+        return aiwin_get_string(aiwin);
+    }
    }
    assert(FALSE);
 }
@@ -662,6 +697,17 @@ win_free(ProfWin* window)
        free(pluginwin->plugin_name);
        break;
    }
+    case WIN_AI:
+    {
+        ProfAiWin* aiwin = (ProfAiWin*)window;
+        assert(aiwin->memcheck == PROFAIWIN_MEMCHECK);
+        buffer_free(aiwin->message_buffer);
+        g_string_free(aiwin->conversation_history, TRUE);
+        if (aiwin->session) {
+            ai_session_unref(aiwin->session);
+        }
+        break;
+    }
    default:
        break;
    }
@@ -802,6 +848,14 @@ win_page_down(ProfWin* window, int scroll_size)
        window->layout->paged = 0;
        window->layout->unread_msg = 0;
    }
+
+    // Non-chat windows have no DB to fetch from, so WIN_SCROLL_REACHED_BOTTOM
+    // never fires; reset paged once the buffer's last line is on screen.
+    if (window->type != WIN_CHAT
+        && (total_rows - *page_start) <= page_space + 1) {
+        window->layout->paged = 0;
+        window->layout->unread_msg = 0;
+    }
 }

 void
@@ -1554,7 +1608,7 @@ win_print_outgoing(ProfWin* window, const char* show_char, const char* const id,
    GDateTime* timestamp = g_date_time_new_now_local();

    const char* myjid = connection_get_fulljid();
-    if (!_win_correct(window, message, id, replace_id, myjid)) {
+    if (!prefs_get_boolean(PREF_CORRECTION_ALLOW) || !_win_correct(window, message, id, replace_id, myjid)) {
        auto_gchar gchar* outgoing_str = prefs_get_string(PREF_OUTGOING_STAMP);
        _win_printf(window, show_char, 0, timestamp, 0, THEME_TEXT_ME, outgoing_str, myjid, id, "%s", message);
    }
@@ -1701,7 +1755,7 @@ win_print_outgoing_with_receipt(ProfWin* window, const char* show_char, const ch
    receipt->received = FALSE;

    const char* myjid = connection_get_fulljid();
-    if (_win_correct(window, message, id, replace_id, myjid)) {
+    if (prefs_get_boolean(PREF_CORRECTION_ALLOW) && _win_correct(window, message, id, replace_id, myjid)) {
        free(receipt); // TODO: probably we should use this in _win_correct()
    } else {
        int y_start_pos = getcury(window->layout->win);
@@ -1755,10 +1809,12 @@ static void
 _win_printf(ProfWin* window, const char* show_char, int pad_indent, GDateTime* timestamp, int flags, theme_item_t theme_item, const char* const display_from, const char* const from_jid, const char* const message_id, const char* const message, ...)
 {

-    /* Prevent printing and buffer update when user is viewing message history [SCROLLING]*/
    if (window->layout->paged && wins_is_current(window)) {
        window->layout->unread_msg++;
-        return;
+        // Skip buffer only when DB can replay the message on scroll-down.
+        if (window->type == WIN_CHAT && log_database_can_recover_messages()) {
+            return;
+        }
    }

    if (timestamp == NULL) {
--- a/src/ui/window_list.c
+++ b/src/ui/window_list.c
@@ -44,9 +44,11 @@

 #include "common.h"
 #include "config/preferences.h"
+#include "log.h"
 #include "config/theme.h"
 #include "plugins/plugins.h"
 #include "ui/ui.h"
+#include "ui/window.h"
 #include "ui/window_list.h"
 #include "xmpp/xmpp.h"
 #include "xmpp/roster_list.h"
@@ -374,6 +376,8 @@ wins_set_current_by_num(int i)
                status_bar_active(1, conswin->type, "console");
            }
        }
+    } else {
+        log_error("Window not found for index %d", i);
    }
 }

@@ -714,6 +718,20 @@ wins_new_vcard(vCard* vcard)
    return newwin;
 }

+ProfWin*
+wins_new_ai(AISession* session)
+{
+    int result = _wins_get_next_available_num(keys);
+    ProfWin* newwin = win_create_ai(session);
+    if (!newwin) {
+        return NULL;
+    }
+    _wins_htable_insert(windows, GINT_TO_POINTER(result), newwin);
+    autocomplete_add(wins_ac, "ai");
+    autocomplete_add(wins_close_ac, "ai");
+    return newwin;
+}
+
 gboolean
 wins_do_notify_remind(void)
 {
@@ -841,7 +859,7 @@ wins_get_prune_wins(void)

    while (curr) {
        ProfWin* window = curr->data;
-        if (win_unread(window) == 0 && window->type != WIN_MUC && window->type != WIN_CONFIG && window->type != WIN_XML && window->type != WIN_CONSOLE) {
+        if (win_unread(window) == 0 && window->type != WIN_MUC && window->type != WIN_CONFIG && window->type != WIN_XML && window->type != WIN_CONSOLE && window->type != WIN_AI) {
            result = g_slist_append(result, window);
        }
        curr = g_list_next(curr);
@@ -849,6 +867,27 @@ wins_get_prune_wins(void)
    return result;
 }

+gboolean
+wins_ai_exists(ProfAiWin* const aiwin)
+{
+    if (!aiwin) {
+        return FALSE;
+    }
+
+    GList* curr = values;
+    while (curr) {
+        ProfWin* window = curr->data;
+        if (window->type == WIN_AI) {
+            if ((ProfAiWin*)window == aiwin) {
+                return TRUE;
+            }
+        }
+        curr = g_list_next(curr);
+    }
+
+    return FALSE;
+}
+
 void
 wins_lost_connection(void)
 {
--- a/src/ui/window_list.h
+++ b/src/ui/window_list.h
@@ -63,6 +63,7 @@ ProfPrivateWin* wins_get_private(const char* const fulljid);
 ProfPluginWin* wins_get_plugin(const char* const tag);
 ProfXMLWin* wins_get_xmlconsole(void);
 ProfVcardWin* wins_get_vcard(void);
+gboolean wins_ai_exists(ProfAiWin* const aiwin);

 void wins_close_plugin(char* tag);

--- a/tests/functionaltests/functionaltests.c
+++ b/tests/functionaltests/functionaltests.c
@@ -59,6 +59,7 @@
 #ifdef HAVE_SQLITE
 #include "test_export_import.h"
 #endif
+#include "test_ai.h"

 /* Macro to wrap each test with setup/teardown functions */
 #define PROF_FUNC_TEST(test)                                \
@@ -66,6 +67,14 @@
        #test, test, init_prof_test, close_prof_test, #test \
    }

+/* AI tests use a custom init that primes stabber via prof_connect so
+ * stbbr_stop() in close_prof_test() doesn't hang on a never-connected
+ * server thread. See ai_init_test() in test_ai.c. */
+#define PROF_FUNC_TEST_AI(test)                          \
+    {                                                    \
+        #test, test, ai_init_test, close_prof_test, #test \
+    }
+
 #ifdef HAVE_SQLITE
 /* Custom init that sets a non-UTC timezone (POSIX TST-3 = UTC+3) */
 static int
@@ -84,9 +93,9 @@ main(int argc, char* argv[])
    int group = 0; /* 0 = all groups */
    if (argc > 1) {
        group = atoi(argv[1]);
-        if (group < 1 || group > 4) {
+        if (group < 1 || group > 5) {
            fprintf(stderr, "Usage: %s [group]\n", argv[0]);
-            fprintf(stderr, "  group: 1-4 to run specific group, or omit for all\n");
+            fprintf(stderr, "  group: 1-5 to run specific group, or omit for all\n");
            return 1;
        }
    }
@@ -307,6 +316,31 @@ main(int argc, char* argv[])
 #endif
    };

+    /* ============================================================
+     * GROUP 5: AI command surface (console + HTTP stub)
+     * Standalone because AI tests don't use stabber's XMPP path; mixing
+     * them with XMPP-bound tests in the same group poisons stabber state
+     * between fixture init/teardown cycles and hangs subsequent
+     * prof_connect() calls.
+     * ============================================================ */
+    const struct CMUnitTest group5_tests[] = {
+        /* Console-only — no network calls */
+        PROF_FUNC_TEST_AI(ai_no_args_shows_help),
+        PROF_FUNC_TEST_AI(ai_providers_lists_defaults),
+        PROF_FUNC_TEST_AI(ai_providers_list_shows_key_status),
+        PROF_FUNC_TEST_AI(ai_set_provider_adds_custom),
+        PROF_FUNC_TEST_AI(ai_set_token_marks_key_set),
+        PROF_FUNC_TEST_AI(ai_start_unknown_provider_errors),
+        PROF_FUNC_TEST_AI(ai_start_without_key_errors),
+        PROF_FUNC_TEST_AI(ai_start_with_key_opens_window),
+        PROF_FUNC_TEST_AI(ai_clear_without_window_errors),
+        PROF_FUNC_TEST_AI(ai_remove_provider_works),
+        PROF_FUNC_TEST_AI(ai_remove_provider_unknown_errors),
+        PROF_FUNC_TEST_AI(ai_set_default_model_updates_provider),
+        PROF_FUNC_TEST_AI(ai_switch_without_window_errors),
+        PROF_FUNC_TEST_AI(ai_bad_subcommand_shows_usage),
+    };
+
    /* Test group registry for easy extension */
    struct
    {
@@ -318,6 +352,7 @@ main(int argc, char* argv[])
        { "Group 2: ExportImport/Receipts", group2_tests, ARRAY_SIZE(group2_tests) },
        { "Group 3: Presence/Disconnect/DBHistory/Message/MUC-DB", group3_tests, ARRAY_SIZE(group3_tests) },
        { "Group 4: Session/MUC/Carbons/MigrationStress", group4_tests, ARRAY_SIZE(group4_tests) },
+        { "Group 5: AI command surface (console only)", group5_tests, ARRAY_SIZE(group5_tests) },
    };
    const int num_groups = ARRAY_SIZE(groups);

--- a/tests/functionaltests/proftest.c
+++ b/tests/functionaltests/proftest.c
@@ -20,7 +20,7 @@
 #include "proftest.h"

 /* Number of parallel test groups for CI builds */
-#define TEST_GROUPS 4
+#define TEST_GROUPS 5

 /* Number of ports reserved per test group; allows skipping TIME_WAIT ports */
 #define PORTS_PER_GROUP 50
--- a/tests/functionaltests/test_ai.c
+++ b/tests/functionaltests/test_ai.c
@@ -0,0 +1,252 @@
+/*
+ * test_ai.c
+ *
+ * Functional tests for the /ai command surface (Tier A — no network).
+ *
+ * These tests interact with profanity through its PTY console and exercise
+ * paths that do not reach libcurl: command parsing, provider registration,
+ * key/setting/default storage, error messages, and AI window creation.
+ *
+ * Tests that require an actual provider HTTP exchange (prompt -> response,
+ * /ai models <provider>, HTTP error envelopes) are out of scope here and
+ * belong in a separate suite backed by a local HTTP stub.
+ */
+
+#include <glib.h>
+#include "prof_cmocka.h"
+#include <stdlib.h>
+#include <string.h>
+
+#include "proftest.h"
+#include "test_ai.h"
+
+int
+ai_init_test(void** state)
+{
+    int ret = init_prof_test(state);
+    if (ret != 0) {
+        return ret;
+    }
+    /* Connect to stabber even though AI tests don't use XMPP. Without this
+     * stabber's accept loop has no client to disconnect on /quit, and
+     * stbbr_stop() in close_prof_test() hangs uninterruptibly when joining
+     * the server thread. */
+    prof_connect();
+    return 0;
+}
+
+void
+ai_no_args_shows_help(void** state)
+{
+    /* `/ai` with no arguments lists the built-in providers and a usage hint.
+     * We don't pin specific provider names — defaults may change. Verify the
+     * header, the "Configured providers" section, that *some* provider line
+     * carries an http(s) URL, and the usage hint. */
+    prof_input("/ai");
+
+    prof_timeout(5);
+    assert_true(prof_output_exact("AI Chat - OpenAI-compatible API client"));
+    assert_true(prof_output_exact("Configured providers:"));
+    assert_true(prof_output_regex("URL: https?://"));
+    assert_true(prof_output_regex("Use '/ai start' to begin a chat"));
+    prof_timeout_reset();
+}
+
+void
+ai_providers_lists_defaults(void** state)
+{
+    /* `/ai providers` (no "list") shows the built-in list with URLs. */
+    prof_input("/ai providers");
+
+    prof_timeout(5);
+    assert_true(prof_output_exact("Available AI providers:"));
+    /* At least one URL line is rendered — exact name agnostic. */
+    assert_true(prof_output_regex("https?://"));
+    prof_timeout_reset();
+}
+
+void
+ai_providers_list_shows_key_status(void** state)
+{
+    /* `/ai providers list` lists each configured provider with key status. */
+    prof_input("/ai providers list");
+
+    prof_timeout(5);
+    assert_true(prof_output_exact("Configured providers:"));
+    /* No tokens have been set yet in this fresh session. */
+    assert_true(prof_output_regex("Key: NOT configured"));
+    prof_timeout_reset();
+}
+
+void
+ai_set_provider_adds_custom(void** state)
+{
+    /* Adding a custom provider should make it appear in /ai providers list. */
+    prof_input("/ai set provider mock http://127.0.0.1:1/");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'mock' configured"));
+    prof_timeout_reset();
+
+    prof_input("/ai providers list");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("mock"));
+    assert_true(prof_output_regex("http://127.0.0.1:1/"));
+    prof_timeout_reset();
+}
+
+void
+ai_set_token_marks_key_set(void** state)
+{
+    /* Use a self-set-up provider so the test doesn't pin a default name. */
+    prof_input("/ai set provider testprov https://example.test/");
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'testprov' configured"));
+    prof_timeout_reset();
+
+    prof_input("/ai set token testprov sk-fake-test-key");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("API token set for provider: testprov"));
+    prof_timeout_reset();
+
+    prof_input("/ai providers list");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Key: configured"));
+    prof_timeout_reset();
+}
+
+void
+ai_start_unknown_provider_errors(void** state)
+{
+    /* Unknown provider name should error out without creating a window. */
+    prof_input("/ai start nope_provider somemodel");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'nope_provider' not found"));
+    prof_timeout_reset();
+}
+
+void
+ai_start_without_key_errors(void** state)
+{
+    /* Self-set-up provider with no token. /ai start must refuse. */
+    prof_input("/ai set provider testprov https://example.test/");
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'testprov' configured"));
+    prof_timeout_reset();
+
+    prof_input("/ai start testprov model-x");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("No API key set for provider 'testprov'"));
+    prof_timeout_reset();
+}
+
+void
+ai_start_with_key_opens_window(void** state)
+{
+    /* Self-set-up provider with token; /ai start opens a WIN_AI window. */
+    prof_input("/ai set provider testprov https://example.test/");
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'testprov' configured"));
+    prof_timeout_reset();
+
+    prof_input("/ai set token testprov sk-fake-test-key");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("API token set for provider: testprov"));
+    prof_timeout_reset();
+
+    prof_input("/ai start testprov model-x");
+
+    prof_timeout(5);
+    /* /ai start switches focus to the new WIN_AI window; the window prints
+     * "AI Chat: <provider>/<model>" as its first line. */
+    assert_true(prof_output_regex("AI Chat: testprov/model-x"));
+    prof_timeout_reset();
+}
+
+void
+ai_clear_without_window_errors(void** state)
+{
+    /* /ai clear from the console (no active AI window) refuses with the
+     * shared "must be in AI chat window" guard used by /ai switch as well. */
+    prof_input("/ai clear");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Must be in an AI chat window"));
+    prof_timeout_reset();
+}
+
+void
+ai_remove_provider_works(void** state)
+{
+    /* Round-trip: add -> verify present -> remove -> verify gone. */
+    prof_input("/ai set provider tmpprov http://127.0.0.1:2/");
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'tmpprov' configured"));
+    prof_timeout_reset();
+
+    prof_input("/ai remove provider tmpprov");
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'tmpprov' removed"));
+    prof_timeout_reset();
+
+    prof_input("/ai remove provider tmpprov");
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'tmpprov' not found"));
+    prof_timeout_reset();
+}
+
+void
+ai_remove_provider_unknown_errors(void** state)
+{
+    /* Removing a provider that was never added must report "not found". */
+    prof_input("/ai remove provider nonexistent_xyz");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'nonexistent_xyz' not found"));
+    prof_timeout_reset();
+}
+
+void
+ai_set_default_model_updates_provider(void** state)
+{
+    /* Setting a default model is acknowledged on the console. */
+    prof_input("/ai set provider testprov https://example.test/");
+    prof_timeout(5);
+    assert_true(prof_output_regex("Provider 'testprov' configured"));
+    prof_timeout_reset();
+
+    prof_input("/ai set default-model testprov model-preview");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Default model for provider 'testprov' set to: model-preview"));
+    prof_timeout_reset();
+}
+
+void
+ai_switch_without_window_errors(void** state)
+{
+    /* /ai switch with no active AI window refuses with the shared
+     * "must be in AI chat window" guard. */
+    prof_input("/ai switch testprov model-x");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Must be in an AI chat window"));
+    prof_timeout_reset();
+}
+
+void
+ai_bad_subcommand_shows_usage(void** state)
+{
+    /* An unknown /ai subcommand should fall through to cons_bad_cmd_usage. */
+    prof_input("/ai not_a_subcommand");
+
+    prof_timeout(5);
+    assert_true(prof_output_regex("Invalid usage, see '/help ai'"));
+    prof_timeout_reset();
+}
--- a/tests/functionaltests/test_ai.h
+++ b/tests/functionaltests/test_ai.h
@@ -0,0 +1,27 @@
+#ifndef __H_FUNC_TEST_AI
+#define __H_FUNC_TEST_AI
+
+/*
+ * AI tests don't exercise the XMPP path, but the test fixture's
+ * stbbr_stop() hangs indefinitely when no client ever connected to
+ * stabber. ai_init_test wraps init_prof_test with a prof_connect()
+ * so stabber sees a graceful disconnect at teardown.
+ */
+int ai_init_test(void** state);
+
+void ai_no_args_shows_help(void** state);
+void ai_providers_lists_defaults(void** state);
+void ai_providers_list_shows_key_status(void** state);
+void ai_set_provider_adds_custom(void** state);
+void ai_set_token_marks_key_set(void** state);
+void ai_start_unknown_provider_errors(void** state);
+void ai_start_without_key_errors(void** state);
+void ai_start_with_key_opens_window(void** state);
+void ai_clear_without_window_errors(void** state);
+void ai_remove_provider_works(void** state);
+void ai_remove_provider_unknown_errors(void** state);
+void ai_set_default_model_updates_provider(void** state);
+void ai_switch_without_window_errors(void** state);
+void ai_bad_subcommand_shows_usage(void** state);
+
+#endif
--- a/tests/unittests/test_ai_client.c
+++ b/tests/unittests/test_ai_client.c
--- a/tests/unittests/test_ai_client.h
+++ b/tests/unittests/test_ai_client.h
@@ -0,0 +1,140 @@
+/*
+ * test_ai_client.h - AI client unit test declarations
+ */
+
+int ai_client_setup(void** state);
+int ai_client_teardown(void** state);
+
+void test_ai_client_init(void** state);
+void test_ai_add_provider(void** state);
+void test_ai_remove_provider(void** state);
+void test_ai_list_providers(void** state);
+void test_ai_set_provider_key(void** state);
+void test_ai_get_provider_key(void** state);
+void test_ai_session_create(void** state);
+void test_ai_session_ref_unref(void** state);
+void test_ai_session_add_message(void** state);
+void test_ai_session_clear_history(void** state);
+void test_ai_session_set_model(void** state);
+void test_ai_json_escape(void** state);
+void test_ai_json_escape_null(void** state);
+void test_ai_json_escape_empty(void** state);
+void test_ai_json_escape_special_chars(void** state);
+void test_ai_json_escape_percent_signs(void** state);
+void test_ai_json_escape_backslash_quote(void** state);
+void test_ai_session_api_key_is_copied(void** state);
+void test_ai_add_provider_update_existing(void** state);
+void test_ai_add_provider_null_name_returns_null(void** state);
+void test_ai_add_provider_null_url_returns_null(void** state);
+void test_ai_session_create_null_provider_returns_null(void** state);
+void test_ai_session_create_null_model_returns_null(void** state);
+void test_ai_session_api_key_null_when_no_key_set(void** state);
+/* Provider autocomplete tests */
+void test_ai_providers_find_forward(void** state);
+void test_ai_providers_find_forward_custom(void** state);
+void test_ai_providers_find_forward_no_match(void** state);
+void test_ai_providers_find_forward_partial_match(void** state);
+void test_ai_providers_find_next(void** state);
+void test_ai_providers_find_previous(void** state);
+void test_ai_providers_find_empty_search_str(void** state);
+void test_ai_providers_find_null_search_str(void** state);
+void test_ai_providers_find_case_insensitive(void** state);
+/* Provider default model and settings tests */
+void test_ai_set_provider_default_model(void** state);
+void test_ai_get_provider_default_model(void** state);
+void test_ai_set_provider_setting(void** state);
+void test_ai_get_provider_setting(void** state);
+/* Model caching tests */
+void test_ai_models_are_fresh_initial(void** state);
+/* Model parsing tests */
+void test_ai_parse_models_openai_format(void** state);
+void test_ai_parse_models_perplexity_format(void** state);
+void test_ai_parse_models_array_format(void** state);
+void test_ai_parse_models_empty_json(void** state);
+void test_ai_parse_models_null_handling(void** state);
+void test_ai_parse_models_escaped_quotes(void** state);
+void test_ai_parse_models_with_whitespace(void** state);
+/* AI autocomplete integration tests */
+void test_ai_start_provider_autocomplete_only_on_exact(void** state);
+void test_ai_providers_find_cycling(void** state);
+
+/* Setup that also initializes prefs for round-trip persistence tests. */
+int ai_client_setup_with_prefs(void** state);
+int ai_client_teardown_with_prefs(void** state);
+
+/* Chat response parser tests (ai_parse_response) */
+void test_ai_parse_response_openai_content(void** state);
+void test_ai_parse_response_perplexity_text(void** state);
+void test_ai_parse_response_text_preferred_over_content(void** state);
+void test_ai_parse_response_escaped_quote(void** state);
+void test_ai_parse_response_newline_escape(void** state);
+void test_ai_parse_response_empty_content(void** state);
+void test_ai_parse_response_missing_field(void** state);
+void test_ai_parse_response_null_input(void** state);
+void test_ai_parse_response_empty_input(void** state);
+void test_ai_parse_response_percent_signs_safe(void** state);
+void test_ai_parse_response_braces_in_content(void** state);
+void test_ai_parse_response_multiline_content(void** state);
+
+/* Error response parser tests (ai_parse_error_message) */
+void test_ai_parse_error_standard_envelope(void** state);
+void test_ai_parse_error_with_escapes(void** state);
+void test_ai_parse_error_no_error_field(void** state);
+void test_ai_parse_error_no_message_field(void** state);
+void test_ai_parse_error_null_input(void** state);
+void test_ai_parse_error_empty_input(void** state);
+void test_ai_parse_error_tab_escape(void** state);
+void test_ai_parse_error_backslash_escape(void** state);
+
+/* Extended JSON escape tests */
+void test_ai_json_escape_backspace(void** state);
+void test_ai_json_escape_formfeed(void** state);
+void test_ai_json_escape_carriage_return(void** state);
+void test_ai_json_escape_all_specials_combined(void** state);
+void test_ai_json_escape_passes_utf8_through(void** state);
+
+/* Autocomplete cycling with many providers */
+void test_ai_providers_find_cycles_through_many(void** state);
+void test_ai_providers_find_previous_walks_backwards(void** state);
+void test_ai_providers_find_wraps_around(void** state);
+
+/* Session edge cases */
+void test_ai_session_add_message_null_session_no_crash(void** state);
+void test_ai_session_add_message_null_role_no_crash(void** state);
+void test_ai_session_add_message_null_content_no_crash(void** state);
+void test_ai_session_history_preserves_large_order(void** state);
+void test_ai_session_clear_empty_history(void** state);
+void test_ai_session_set_model_null_keeps_old(void** state);
+void test_ai_session_unref_null_no_crash(void** state);
+void test_ai_session_ref_null_returns_null(void** state);
+
+/* Provider edge cases */
+void test_ai_get_provider_null_returns_null(void** state);
+void test_ai_remove_provider_null_returns_false(void** state);
+void test_ai_remove_provider_twice_second_fails(void** state);
+void test_ai_provider_survives_via_session_after_removal(void** state);
+
+/* Settings edge cases */
+void test_ai_settings_multiple_keys_independent(void** state);
+void test_ai_settings_get_missing_returns_null(void** state);
+void test_ai_settings_isolated_between_providers(void** state);
+
+/* Model parsing edge cases */
+void test_ai_parse_models_data_not_array(void** state);
+void test_ai_parse_models_empty_data_array(void** state);
+void test_ai_parse_models_id_outside_data_ignored(void** state);
+void test_ai_parse_models_multiple_models(void** state);
+
+/* Prefs round-trip tests */
+void test_ai_prefs_round_trip_api_key(void** state);
+void test_ai_prefs_round_trip_remove_key(void** state);
+void test_ai_prefs_multiple_providers_persist(void** state);
+
+/* Autocomplete deeper coverage */
+void test_ai_providers_find_after_remove_skips_removed(void** state);
+
+/* Reset hook + persistence */
+void test_ai_providers_reset_ac_restarts_cycle(void** state);
+void test_ai_add_provider_persisted_across_init(void** state);
+void test_ai_remove_provider_persisted_across_init(void** state);
+void test_ai_models_persisted_across_init(void** state);
--- a/tests/unittests/test_cmd_otr.c
+++ b/tests/unittests/test_cmd_otr.c
@@ -69,7 +69,7 @@ cmd_otr_log_on_shows_warning_when_chlog_disabled(void** state)
    prefs_set_boolean(PREF_CHLOG, FALSE);

    expect_cons_show("OTR messages will be logged as plaintext.");
-    expect_cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+    expect_cons_show("Chat logging is currently disabled, use '/history on' to enable.");

    gboolean result = cmd_otr_log(NULL, CMD_OTR, args);
    assert_true(result);
@@ -115,7 +115,7 @@ cmd_otr_log_redact_shows_warning_when_chlog_disabled(void** state)
    prefs_set_boolean(PREF_CHLOG, FALSE);

    expect_cons_show("OTR messages will be logged as '[redacted]'.");
-    expect_cons_show("Chat logging is currently disabled, use '/logging chat on' to enable.");
+    expect_cons_show("Chat logging is currently disabled, use '/history on' to enable.");

    gboolean result = cmd_otr_log(NULL, CMD_OTR, args);
    assert_true(result);
--- a/tests/unittests/ui/stub_ai.c
+++ b/tests/unittests/ui/stub_ai.c
@@ -0,0 +1,41 @@
+/*
+ * stub_ai.c - Stubs for AI window functions
+ */
+
+#include "ui/window.h"
+#include "ai/ai_client.h"
+
+ProfWin*
+win_create_ai(AISession* session)
+{
+    /* Return NULL to simulate failure in unit tests */
+    (void)session;
+    return NULL;
+}
+
+void
+aiwin_display_response(ProfAiWin* win, const char* response)
+{
+    /* Stub: do nothing */
+    (void)win;
+    (void)response;
+}
+
+void
+aiwin_display_error(ProfAiWin* win, const char* error_msg)
+{
+    /* Stub: do nothing */
+    (void)win;
+    (void)error_msg;
+}
+
+void
+win_print_outgoing(ProfWin* window, const char* show_char, const char* const id, const char* const replace_id, const char* const message)
+{
+    /* Stub: do nothing */
+    (void)window;
+    (void)show_char;
+    (void)id;
+    (void)replace_id;
+    (void)message;
+}
--- a/tests/unittests/unittests.c
+++ b/tests/unittests/unittests.c
@@ -36,6 +36,7 @@
 #include "test_form.h"
 #include "test_callbacks.h"
 #include "test_plugins_disco.h"
+#include "test_ai_client.h"
 #include "test_database_export.h"
 #include "test_database_stress.h"

@@ -659,6 +660,128 @@ main(int argc, char* argv[])
        cmocka_unit_test_setup_teardown(test_cmd_force_encryption_invalid_policy, load_preferences, close_preferences),
        cmocka_unit_test_setup_teardown(test_allow_unencrypted_message_invalid_mode, load_preferences, close_preferences),

+        // AI client tests
+        cmocka_unit_test_setup_teardown(test_ai_client_init, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_add_provider, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_remove_provider, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_list_providers, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_set_provider_key, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_get_provider_key, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_create, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_ref_unref, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_add_message, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_clear_history, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_set_model, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_api_key_is_copied, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_add_provider_update_existing, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_add_provider_null_name_returns_null, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_add_provider_null_url_returns_null, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_create_null_provider_returns_null, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_create_null_model_returns_null, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_api_key_null_when_no_key_set, ai_client_setup, ai_client_teardown),
+        /* Provider autocomplete tests */
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_forward, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_forward_custom, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_forward_no_match, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_forward_partial_match, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_next, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_previous, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_empty_search_str, ai_client_setup, ai_client_teardown),
+        /* SIGSEGV on ai_providers_find(NULL, ...) — see test body and
+         * AI_AUTOCOMPLETE_POSSIBLE_ISSUES.md, issue 1. Re-enable once the
+         * NULL-search path is guarded. */
+        /* cmocka_unit_test_setup_teardown(test_ai_providers_find_null_search_str, ai_client_setup, ai_client_teardown), */
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_case_insensitive, ai_client_setup, ai_client_teardown),
+        /* Provider default model and settings tests */
+        cmocka_unit_test_setup_teardown(test_ai_set_provider_default_model, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_get_provider_default_model, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_set_provider_setting, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_get_provider_setting, ai_client_setup, ai_client_teardown),
+        /* Model caching tests */
+        cmocka_unit_test_setup_teardown(test_ai_models_are_fresh_initial, ai_client_setup, ai_client_teardown),
+        /* Model parsing tests */
+        cmocka_unit_test(test_ai_parse_models_openai_format),
+        cmocka_unit_test(test_ai_parse_models_perplexity_format),
+        cmocka_unit_test(test_ai_parse_models_array_format),
+        cmocka_unit_test(test_ai_parse_models_empty_json),
+        cmocka_unit_test(test_ai_parse_models_null_handling),
+        cmocka_unit_test(test_ai_parse_models_escaped_quotes),
+        cmocka_unit_test(test_ai_parse_models_with_whitespace),
+        /* AI autocomplete integration tests */
+        cmocka_unit_test_setup_teardown(test_ai_start_provider_autocomplete_only_on_exact, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_cycling, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test(test_ai_json_escape),
+        cmocka_unit_test(test_ai_json_escape_null),
+        cmocka_unit_test(test_ai_json_escape_empty),
+        cmocka_unit_test(test_ai_json_escape_special_chars),
+        cmocka_unit_test(test_ai_json_escape_percent_signs),
+        cmocka_unit_test(test_ai_json_escape_backslash_quote),
+        /* Chat response parser */
+        cmocka_unit_test(test_ai_parse_response_openai_content),
+        cmocka_unit_test(test_ai_parse_response_perplexity_text),
+        cmocka_unit_test(test_ai_parse_response_text_preferred_over_content),
+        cmocka_unit_test(test_ai_parse_response_escaped_quote),
+        cmocka_unit_test(test_ai_parse_response_newline_escape),
+        cmocka_unit_test(test_ai_parse_response_empty_content),
+        cmocka_unit_test(test_ai_parse_response_missing_field),
+        cmocka_unit_test(test_ai_parse_response_null_input),
+        cmocka_unit_test(test_ai_parse_response_empty_input),
+        cmocka_unit_test(test_ai_parse_response_percent_signs_safe),
+        cmocka_unit_test(test_ai_parse_response_braces_in_content),
+        cmocka_unit_test(test_ai_parse_response_multiline_content),
+        /* Error response parser */
+        cmocka_unit_test(test_ai_parse_error_standard_envelope),
+        cmocka_unit_test(test_ai_parse_error_with_escapes),
+        cmocka_unit_test(test_ai_parse_error_no_error_field),
+        cmocka_unit_test(test_ai_parse_error_no_message_field),
+        cmocka_unit_test(test_ai_parse_error_null_input),
+        cmocka_unit_test(test_ai_parse_error_empty_input),
+        cmocka_unit_test(test_ai_parse_error_tab_escape),
+        cmocka_unit_test(test_ai_parse_error_backslash_escape),
+        /* Extended JSON escape coverage */
+        cmocka_unit_test(test_ai_json_escape_backspace),
+        cmocka_unit_test(test_ai_json_escape_formfeed),
+        cmocka_unit_test(test_ai_json_escape_carriage_return),
+        cmocka_unit_test(test_ai_json_escape_all_specials_combined),
+        cmocka_unit_test(test_ai_json_escape_passes_utf8_through),
+        /* Autocomplete cycling */
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_cycles_through_many, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_previous_walks_backwards, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_wraps_around, ai_client_setup, ai_client_teardown),
+        /* Session edge cases */
+        cmocka_unit_test_setup_teardown(test_ai_session_add_message_null_session_no_crash, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_add_message_null_role_no_crash, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_add_message_null_content_no_crash, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_history_preserves_large_order, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_clear_empty_history, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_set_model_null_keeps_old, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_unref_null_no_crash, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_session_ref_null_returns_null, ai_client_setup, ai_client_teardown),
+        /* Provider edge cases */
+        cmocka_unit_test_setup_teardown(test_ai_get_provider_null_returns_null, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_remove_provider_null_returns_false, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_remove_provider_twice_second_fails, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_provider_survives_via_session_after_removal, ai_client_setup, ai_client_teardown),
+        /* Settings */
+        cmocka_unit_test_setup_teardown(test_ai_settings_multiple_keys_independent, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_settings_get_missing_returns_null, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_settings_isolated_between_providers, ai_client_setup, ai_client_teardown),
+        /* Model parsing edge cases */
+        cmocka_unit_test_setup_teardown(test_ai_parse_models_data_not_array, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_parse_models_empty_data_array, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_parse_models_id_outside_data_ignored, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_parse_models_multiple_models, ai_client_setup, ai_client_teardown),
+        /* Prefs round-trip (uses prefs+ai setup) */
+        cmocka_unit_test_setup_teardown(test_ai_prefs_round_trip_api_key, ai_client_setup_with_prefs, ai_client_teardown_with_prefs),
+        cmocka_unit_test_setup_teardown(test_ai_prefs_round_trip_remove_key, ai_client_setup_with_prefs, ai_client_teardown_with_prefs),
+        cmocka_unit_test_setup_teardown(test_ai_prefs_multiple_providers_persist, ai_client_setup_with_prefs, ai_client_teardown_with_prefs),
+        /* Autocomplete deeper coverage */
+        cmocka_unit_test_setup_teardown(test_ai_providers_find_after_remove_skips_removed, ai_client_setup, ai_client_teardown),
+        /* Reset hook + persistence */
+        cmocka_unit_test_setup_teardown(test_ai_providers_reset_ac_restarts_cycle, ai_client_setup, ai_client_teardown),
+        cmocka_unit_test_setup_teardown(test_ai_add_provider_persisted_across_init, ai_client_setup_with_prefs, ai_client_teardown_with_prefs),
+        cmocka_unit_test_setup_teardown(test_ai_remove_provider_persisted_across_init, ai_client_setup_with_prefs, ai_client_teardown_with_prefs),
+        cmocka_unit_test_setup_teardown(test_ai_models_persisted_across_init, ai_client_setup_with_prefs, ai_client_teardown_with_prefs),
        // Flatfile export/import round-trip
        cmocka_unit_test(test_ff_roundtrip_simple_chat),
        cmocka_unit_test(test_ff_roundtrip_with_all_metadata),
--- a/tests/unittests/xmpp/stub_xmpp.c
+++ b/tests/unittests/xmpp/stub_xmpp.c
@@ -104,6 +104,12 @@ connection_create_uuid(void)
    return NULL;
 }

+char*
+connection_create_stanza_id(void)
+{
+    return NULL;
+}
+
 void
 connection_free_uuid(char* uuid)
 {
Author	SHA1	Message	Date
Jabber Developer	53cdf488b6	fix(ai): resolve UAF and require provider argument instead of defaults All checks were successful CI Code / Check spelling (pull_request) Successful in 18s Details CI Code / Check coding style (pull_request) Successful in 50s Details CI Code / Linux (debian) (pull_request) Successful in 4m57s Details CI Code / Linux (ubuntu) (pull_request) Successful in 5m12s Details CI Code / Code Coverage (pull_request) Successful in 6m48s Details CI Code / Linux (arch) (pull_request) Successful in 11m9s Details CI Code / Check spelling (push) Successful in 16s Details CI Code / Check coding style (push) Successful in 33s Details CI Code / Code Coverage (push) Successful in 2m41s Details CI Code / Linux (debian) (push) Successful in 4m40s Details CI Code / Linux (ubuntu) (push) Successful in 4m53s Details CI Code / Linux (arch) (push) Successful in 10m41s Details This enforces explicit provider specification. Automatic fallback to preferences or "openai" is removed. Users must now pass the provider name explicitly. default provider functionality was improperly removed and implemented (some parts still remain intact), leading to UAF.	2026-05-19 16:42:22 +00:00
Jabber Developer	4776c1f1ec	fix(ai): properly decode \uXXXX JSON escape sequences as UTF-8 Add helper functions to parse hex digits and encode UTF-8 characters. Update buffer allocation to account for UTF-8 expansion and implement full surrogate pair support for characters outside the BMP. Previously, \uXXXX sequences were passed through verbatim; they are now correctly decoded into proper UTF-8 strings.	2026-05-19 16:40:21 +00:00
Jabber Developer	a3a45ad477	fix(ui): preserve messages in non-chat windows while scrolled All checks were successful CI Code / Check spelling (push) Successful in 17s Details CI Code / Check coding style (push) Successful in 31s Details CI Code / Code Coverage (push) Successful in 2m43s Details CI Code / Linux (debian) (push) Successful in 4m33s Details CI Code / Linux (ubuntu) (push) Successful in 4m48s Details CI Code / Linux (arch) (push) Successful in 5m36s Details _win_printf dropped buffer append and render for any window in paged state, but only WIN_CHAT can recover lost messages via chatwin_db_history() on scroll-down. WIN_MUC, WIN_PRIVATE and WIN_AI have no such fallback, so incoming and outgoing messages were silently lost when the user was viewing history. Introduce log_database_can_recover_messages() to check whether the DB backend can replay messages (returns FALSE when PREF_DBLOG is "off", "redact", or no backend is active). Gate the WIN_CHAT early return in _win_printf() on this check: when recovery is impossible, fall through and append to the buffer so messages remain visible on scroll-down. Add a buffer-bottom reset in win_page_down() for non-chat windows. WIN_SCROLL_REACHED_BOTTOM is only set on the is_chat DB branch, so non-chat windows never clear paged on their own; reset paged and unread_msg when the last line of the buffer reaches the screen. Add missing scroll rendering for AI windows in the title bar draw function, ensuring AI windows display their scrolled state consistently with other window types. Remove the manual paged/unread_msg reset before printing the user message in cl_ev_send_ai_msg() -- it was a local workaround for the same drop and is no longer needed.	2026-05-16 15:47:49 +00:00
Jabber Developer	2952466abd	feat(history): consolidate logging controls and add dbbackend statusbar indicator All checks were successful CI Code / Check spelling (push) Successful in 18s Details CI Code / Check coding style (push) Successful in 33s Details CI Code / Code Coverage (push) Successful in 2m42s Details CI Code / Linux (debian) (push) Successful in 4m40s Details CI Code / Linux (ubuntu) (push) Successful in 4m53s Details CI Code / Linux (arch) (push) Successful in 5m38s Details Deprecate /logging command in favor of /history for chat logging control. /history off now stops persistence (sets PREF_DBLOG=off + PREF_CHLOG=false) in addition to hiding history on open. /history on restores persistence (re-enabling PREF_DBLOG=on if it was off) and PREF_CHLOG. statusbar - PREF_STATUSBAR_SHOW_DBBACKEND (default ON) gates the [sqlite] / [flatfile] indicator; toggle via "/statusbar show\|hide dbbackend" /history off\|on - "/history off" now stops persistence as well as hiding history on open: sets PREF_DBLOG=off + PREF_CHLOG=false in addition to PREF_HISTORY=false - "/history on" restores persistence (re-enabling PREF_DBLOG=on if it was off) and PREF_CHLOG, in addition to PREF_HISTORY=true /logging - Deprecated /logging command. It now prints a single notice pointing to /history; CMD_PREAMBLE trimmed (min_args=0, no setting_func, syntax/args/examples dropped); subcommand 'group' removed from autocomplete - All "use '/logging chat on' to enable" hints replaced with "/history on" across /omemo-log, /pgp-log, /otr-log, and /ox-log /privacy logging - Single-pass validation across {on, off, redact, flatfile} - Backend-switching values (on, flatfile) now take effect immediately when connected (via log_database_switch_backend), matching "/history switch" behaviour; off/redact only flip pref bits and keep the live backend open /correction off - win_print_outgoing and win_print_outgoing_with_receipt now gate _win_correct on PREF_CORRECTION_ALLOW, matching incoming-msg paths. Previously a peer's correction reflected via XEP-0280 carbons (or the user's own /correct invocation) was applied in-buffer regardless of the pref console - Drop orphaned /logging chat reference from cons_privacy_setting() - Delete cons_logging_setting() and remove its call from cons_show_log_prefs() autocomplete - Add dbbackend to statusbar_show_ac - Remove logging_ac entries for chat/group subcommands - Remove _logging_autocomplete() param handler for chat subcommand database_flatfile - Two local g_strndup allocations switched from char* with manual g_free to auto_gchar gchar* for automatic cleanup tests - Update test_cmd_otr.c to expect new /history-on warning messages @author: jabber.developer2 <jabber.developer2@jabber.space>	2026-05-16 15:33:42 +00:00
Jabber Developer	06b80bc89a	fix(ai): fix memory leak of local provider in error paths All checks were successful CI Code / Check spelling (pull_request) Successful in 17s Details CI Code / Check coding style (pull_request) Successful in 31s Details CI Code / Code Coverage (pull_request) Successful in 2m44s Details CI Code / Linux (debian) (pull_request) Successful in 4m42s Details CI Code / Linux (ubuntu) (pull_request) Successful in 4m56s Details CI Code / Linux (arch) (pull_request) Successful in 5m49s Details CI Code / Check spelling (push) Successful in 16s Details CI Code / Check coding style (push) Successful in 31s Details CI Code / Code Coverage (push) Successful in 2m42s Details CI Code / Linux (debian) (push) Successful in 4m45s Details CI Code / Linux (ubuntu) (push) Successful in 4m57s Details CI Code / Linux (arch) (push) Successful in 5m50s Details Add missing unref calls for local_provider in _ai_request_thread error handling to prevent memory leaks.	2026-05-15 14:14:11 +00:00
Jabber Developer	f9e0ba9630	fix(ai): fix memory leaks in ai session handling All checks were successful CI Code / Check spelling (pull_request) Successful in 19s Details CI Code / Check coding style (pull_request) Successful in 38s Details CI Code / Code Coverage (pull_request) Successful in 2m55s Details CI Code / Linux (debian) (pull_request) Successful in 4m39s Details CI Code / Linux (ubuntu) (pull_request) Successful in 4m51s Details CI Code / Linux (arch) (pull_request) Successful in 5m34s Details Add missing `ai_session_unref` calls to prevent memory leaks. - In `_ai_request_thread`, release session on error paths and after successful processing. - In `cmd_ai_start`, release reference after passing ownership to the AI window.	2026-05-15 11:58:25 +00:00
Jabber Developer	7469f31c78	fix(ai): fix memory leaks in _ai_request_thread All checks were successful CI Code / Check spelling (pull_request) Successful in 17s Details CI Code / Check coding style (pull_request) Successful in 34s Details CI Code / Code Coverage (pull_request) Successful in 2m43s Details CI Code / Linux (debian) (pull_request) Successful in 4m45s Details CI Code / Linux (ubuntu) (pull_request) Successful in 4m55s Details CI Code / Linux (arch) (pull_request) Successful in 5m45s Details Replace manual g_free calls with auto_gchar for local_provider_name, local_model, local_api_key, and response_data to ensure automatic cleanup and prevent memory leaks.	2026-05-15 11:29:10 +00:00
Jabber Developer	5e329c77e1	fix(ai): fix memory leak in AI message stanza ID All checks were successful CI Code / Check spelling (pull_request) Successful in 17s Details CI Code / Check coding style (pull_request) Successful in 32s Details CI Code / Code Coverage (pull_request) Successful in 2m43s Details CI Code / Linux (debian) (pull_request) Successful in 4m41s Details CI Code / Linux (ubuntu) (pull_request) Successful in 4m55s Details CI Code / Linux (arch) (pull_request) Successful in 5m44s Details Store stanza ID in an auto_gchar variable to ensure automatic memory cleanup.	2026-05-15 08:55:56 +00:00
Jabber Developer	9e5dfb14f8	feat(ai): add AI client with multi-provider chat support All checks were successful CI Code / Check spelling (push) Successful in 18s Details CI Code / Check coding style (push) Successful in 29s Details CI Code / Code Coverage (push) Successful in 2m44s Details CI Code / Linux (debian) (push) Successful in 4m46s Details CI Code / Linux (ubuntu) (push) Successful in 4m59s Details CI Code / Linux (arch) (push) Successful in 5m56s Details Add an AI client module that integrates with OpenAI-compatible API providers (OpenAI, Perplexity, and custom endpoints) to provide AI-assisted chat within CProof. Users can start sessions with /ai start, send prompts, receive responses in a dedicated AI window, switch between providers and models, and manage API keys — all with tab-completion. Providers are configured via /ai set commands with per-provider API keys, endpoints, default models, and custom settings. Two default providers (openai, perplexity) are seeded on first use. Provider state persists in [ai/<name>] sections of the preferences keyfile with automatic migration from the previous flat-key format. The /ai command integrates into the existing command system with 8 subcommands covering provider management, session lifecycle, model fetching, and conversation clearing. Autocomplete uses the standard flat prefix-matching chain for reliable tab-completion at every nesting level. A new ProfAiWin window type is added to the window system. Architecture: Async design: HTTP requests run on a background thread (pthread) to avoid blocking the ncurses UI loop; results are displayed on the main thread via direct function calls Thread safety: AIProvider and AISession use atomic ref-counting and mutex-protected session state; the request thread snapshots all session data before making the HTTP call Window validation: wins_ai_exists() prevents use-after-free when the user closes the AI window during an in-flight HTTP request (~60s) Privacy: store:false is sent with every request to prevent providers from persisting conversations or using them for training Response size limit: 10MB cap with immediate curl abort via CURL_WRITEFUNC_ERROR to prevent OOM JSON parsing uses unified helpers for both chat responses and error envelopes with consistent escape decoding. The response parser tries Perplexity /v1/responses "text" field first, then falls back to OpenAI "content". Error parsing extracts provider error.message from the standard envelope format. Model parsing handles multiple API response formats (OpenAI list, Perplexity, array) including edge cases. Tests include 470+ lines of unit tests covering provider management, session lifecycle, JSON parsing (multiple formats), autocomplete cycling, and error handling, plus functional tests for /ai command dispatch. A stub_ai.c module isolates unit tests from UI dependencies.	2026-05-15 02:21:54 +00:00