+
diff --git a/src/config/llm.rs b/src/config/llm.rs
index 9605eb27..bd56bec8 100644
--- a/src/config/llm.rs
+++ b/src/config/llm.rs
@@ -342,8 +342,14 @@ impl LlmConfig {
}
Some(creds.token)
} else if let Some(env_var) = api_key_env {
- // Resolve API key from env (including secrets store overlay)
- optional_env(env_var)?.map(SecretString::from)
+ // Resolve API key: env var (including secrets store overlay) > settings override
+ optional_env(env_var)?.map(SecretString::from).or_else(|| {
+ settings
+ .llm_builtin_overrides
+ .get(backend)
+ .and_then(|o| o.api_key.as_ref())
+ .map(|k| SecretString::from(k.clone()))
+ })
} else {
None
};
@@ -392,8 +398,16 @@ impl LlmConfig {
});
}
- // Resolve model
- let model = Self::resolve_model(model_env, settings, default_model)?;
+ // Resolve model: env var > selected_model (/model command) > per-provider override > registry default
+ let model = optional_env(model_env)?
+ .or_else(|| settings.selected_model.clone())
+ .or_else(|| {
+ settings
+ .llm_builtin_overrides
+ .get(backend)
+ .and_then(|o| o.model.clone())
+ })
+ .unwrap_or_else(|| default_model.to_string());
// Resolve extra headers
let extra_headers = if let Some(env_var) = extra_headers_env {
@@ -1193,4 +1207,95 @@ mod tests {
std::env::remove_var("LLM_BACKEND");
}
}
+
+ #[test]
+ fn builtin_override_model_used_when_no_selected_model() {
+ let _guard = ENV_MUTEX.lock().expect("env mutex poisoned");
+ // SAFETY: Under ENV_MUTEX.
+ unsafe {
+ std::env::remove_var("LLM_BACKEND");
+ std::env::remove_var("GROQ_MODEL");
+ }
+
+ let mut overrides = std::collections::HashMap::new();
+ overrides.insert(
+ "groq".to_string(),
+ crate::settings::LlmBuiltinOverride { api_key: None, model: Some("llama-3.1-8b-instant".to_string()) },
+ );
+ let settings = Settings {
+ llm_backend: Some("groq".to_string()),
+ llm_builtin_overrides: overrides,
+ ..Default::default()
+ };
+
+ let cfg = LlmConfig::resolve(&settings).expect("resolve should succeed");
+ let provider = cfg.provider.expect("provider config should be present");
+ assert_eq!(
+ provider.model, "llama-3.1-8b-instant",
+ "builtin override model should be used when selected_model is unset"
+ );
+ }
+
+ #[test]
+ fn selected_model_takes_priority_over_builtin_override_model() {
+ let _guard = ENV_MUTEX.lock().expect("env mutex poisoned");
+ // SAFETY: Under ENV_MUTEX.
+ unsafe {
+ std::env::remove_var("LLM_BACKEND");
+ std::env::remove_var("GROQ_MODEL");
+ }
+
+ let mut overrides = std::collections::HashMap::new();
+ overrides.insert(
+ "groq".to_string(),
+ crate::settings::LlmBuiltinOverride { api_key: None, model: Some("llama-3.1-8b-instant".to_string()) },
+ );
+ let settings = Settings {
+ llm_backend: Some("groq".to_string()),
+ selected_model: Some("llama-3.3-70b-versatile".to_string()),
+ llm_builtin_overrides: overrides,
+ ..Default::default()
+ };
+
+ let cfg = LlmConfig::resolve(&settings).expect("resolve should succeed");
+ let provider = cfg.provider.expect("provider config should be present");
+ assert_eq!(
+ provider.model, "llama-3.3-70b-versatile",
+ "selected_model (/model command) must take priority over builtin override"
+ );
+ }
+
+ #[test]
+ fn builtin_override_api_key_used_when_no_env_var() {
+ let _guard = ENV_MUTEX.lock().expect("env mutex poisoned");
+ // SAFETY: Under ENV_MUTEX.
+ unsafe {
+ std::env::remove_var("LLM_BACKEND");
+ std::env::remove_var("GROQ_API_KEY");
+ std::env::remove_var("GROQ_MODEL");
+ }
+
+ let mut overrides = std::collections::HashMap::new();
+ overrides.insert(
+ "groq".to_string(),
+ crate::settings::LlmBuiltinOverride {
+ api_key: Some("gsk_test_key".to_string()),
+ model: Some("llama-3.3-70b-versatile".to_string()),
+ },
+ );
+ let settings = Settings {
+ llm_backend: Some("groq".to_string()),
+ llm_builtin_overrides: overrides,
+ ..Default::default()
+ };
+
+ let cfg = LlmConfig::resolve(&settings).expect("resolve should succeed");
+ let provider = cfg.provider.expect("provider config should be present");
+ use secrecy::ExposeSecret as _;
+ let key = provider.api_key.expect("api_key should be set from builtin override");
+ assert_eq!(
+ key.expose_secret(), "gsk_test_key",
+ "builtin override api_key should be used when env var is absent"
+ );
+ }
}
diff --git a/src/llm/rig_adapter.rs b/src/llm/rig_adapter.rs
index 5ca4fec0..7b336def 100644
--- a/src/llm/rig_adapter.rs
+++ b/src/llm/rig_adapter.rs
@@ -1370,4 +1370,68 @@ mod tests {
// Should be 2 separate User messages (text user + tool result user)
assert_eq!(history.len(), 2);
}
+
+ /// Empty user messages (e.g. after thinking-tag stripping) must be skipped.
+ /// Strict providers like Kimi return 400 when "content": "" is sent.
+ #[test]
+ fn test_empty_user_message_is_skipped() {
+ let empty = ChatMessage::user("");
+ let non_empty = ChatMessage::user("hello");
+ let messages = vec![empty, non_empty];
+ let (_preamble, history) = convert_messages(&messages);
+
+ assert_eq!(history.len(), 1, "empty user message must be dropped");
+ match &history[0] {
+ RigMessage::User { content } => {
+ assert_eq!(content.len(), 1);
+ let first = content.iter().next().expect("one content item");
+ match first {
+ UserContent::Text(t) => assert_eq!(t.text, "hello"),
+ other => panic!("expected Text, got {:?}", other),
+ }
+ }
+ other => panic!("expected User message, got {:?}", other),
+ }
+ }
+
+ /// Empty assistant messages (e.g. after thinking-tag stripping) must be skipped.
+ #[test]
+ fn test_empty_assistant_message_is_skipped() {
+ let empty_asst = ChatMessage {
+ role: crate::llm::Role::Assistant,
+ content: String::new(),
+ tool_calls: None,
+ tool_call_id: None,
+ name: None,
+ content_parts: vec![],
+ };
+ let non_empty = ChatMessage::user("hi");
+ let messages = vec![empty_asst, non_empty];
+ let (_preamble, history) = convert_messages(&messages);
+
+ assert_eq!(history.len(), 1, "empty assistant message must be dropped");
+ assert!(matches!(history[0], RigMessage::User { .. }));
+ }
+
+ /// A conversation mixing normal and empty messages: only non-empty ones survive.
+ #[test]
+ fn test_mixed_empty_and_non_empty_messages_filtered_correctly() {
+ let user1 = ChatMessage::user("first");
+ let empty_asst = ChatMessage {
+ role: crate::llm::Role::Assistant,
+ content: String::new(),
+ tool_calls: None,
+ tool_call_id: None,
+ name: None,
+ content_parts: vec![],
+ };
+ let user2 = ChatMessage::user("");
+ let asst = ChatMessage::assistant("response");
+ let messages = vec![user1, empty_asst, user2, asst];
+ let (_preamble, history) = convert_messages(&messages);
+
+ assert_eq!(history.len(), 2, "only non-empty messages should survive");
+ assert!(matches!(history[0], RigMessage::User { .. }));
+ assert!(matches!(history[1], RigMessage::Assistant { .. }));
+ }
}
diff --git a/src/settings.rs b/src/settings.rs
index 6e355a8a..1750f697 100644
--- a/src/settings.rs
+++ b/src/settings.rs
@@ -3,6 +3,7 @@
//! Stores user preferences in ~/.ironclaw/settings.json.
//! Settings are loaded with env var > settings.json > default priority.
+use std::collections::HashMap;
use std::path::PathBuf;
use serde::{Deserialize, Serialize};
@@ -32,6 +33,20 @@ pub struct CustomLlmProviderSettings {
pub builtin: bool,
}
+/// Per-provider overrides for built-in LLM providers (API key and/or model).
+///
+/// Stored as `llm_builtin_overrides` in the settings store, keyed by provider ID
+/// (e.g. `"openai"`, `"gemini"`). Resolved at startup during `LlmConfig::resolve()`.
+#[derive(Debug, Clone, Default, Serialize, Deserialize)]
+pub struct LlmBuiltinOverride {
+ /// API key override (takes precedence over env var).
+ #[serde(default, skip_serializing_if = "Option::is_none")]
+ pub api_key: Option
,
+ /// Default model override (takes precedence over global `selected_model`).
+ #[serde(default, skip_serializing_if = "Option::is_none")]
+ pub model: Option,
+}
+
/// User settings persisted to disk.
#[derive(Debug, Clone, Serialize, Deserialize, Default)]
pub struct Settings {
@@ -86,6 +101,10 @@ pub struct Settings {
#[serde(default)]
pub llm_custom_providers: Vec,
+ /// Per-provider overrides for built-in providers (API key and/or model).
+ #[serde(default)]
+ pub llm_builtin_overrides: HashMap,
+
/// Ollama base URL (when llm_backend = "ollama").
#[serde(default)]
pub ollama_base_url: Option,