Max Turnの実装

crates/llm-worker-persistence/src/session.rs

@@ -320,6 +320,7 @@ impl<C: LlmClient, St: Store> Session<C, St> {
         let outcome = match result {
             Ok(WorkerResult::Finished) => Outcome::Finished,
             Ok(WorkerResult::Paused) => Outcome::Paused,
+            Ok(WorkerResult::LimitReached) => Outcome::LimitReached,
             Err(e) => Outcome::Error {
                 message: e.to_string(),
             },

crates/llm-worker-persistence/src/session_log.rs

@@ -67,6 +67,7 @@ pub enum LogEntry {
 pub enum Outcome {
     Finished,
     Paused,
+    LimitReached,
     Error { message: String },
 }
 

crates/llm-worker/README.md

@@ -20,3 +20,4 @@ LLM との対話を管理する低レベル基盤クレート。会話履歴、
 - `timeline` — イベントストリームのディスパッチ（`Handler` トレイト、各ブロックコレクター）
 - `event` — ストリーミングイベント型（`Event`, `BlockStart`, `BlockDelta` など）
 - `state` — 型状態パターンによるキャッシュ保護（`Mutable` / `CacheLocked`）
+cratesの整理Add READMEsRE to all crates@@

crates/llm-worker/examples/worker_cancel_demo.rs

@@ -49,6 +49,9 @@ async fn main() -> Result<(), Box<dyn std::error::Error>> {
             Ok(WorkerResult::Paused) => {
                 println!("⏸️  Task paused");
             }
+            Ok(WorkerResult::LimitReached) => {
+                println!("🔒 Turn limit reached");
+            }
             Err(e) => {
                 println!("❌ Task error: {}", e);
             }

crates/llm-worker/src/worker.rs

@@ -84,6 +84,8 @@ pub enum WorkerResult {
     Finished,
     /// Paused (can be resumed)
     Paused,
+    /// Turn limit reached (max_turns exceeded)
+    LimitReached,
 }
 
 /// Internal: tool execution result
@@ -179,6 +181,8 @@ pub struct Worker<C: LlmClient, S: WorkerState = Mutable> {
     locked_prefix_len: usize,
     /// Turn count
     turn_count: usize,
+    /// Maximum number of turns (None = unlimited)
+    max_turns: Option<u32>,
     /// Turn notification callbacks
     turn_notifiers: Vec<Box<dyn TurnNotifier>>,
     /// Request configuration (max_tokens, temperature, etc.)
@@ -1097,6 +1101,15 @@ impl<C: LlmClient, S: WorkerState> Worker<C, S> {
                     return Err(err);
                 }
             }
+
+            // Check turn limit (after assistant items and tool results are in history)
+            if let Some(max) = self.max_turns {
+                if self.turn_count >= max as usize {
+                    info!(turn_count = self.turn_count, max_turns = max, "Turn limit reached");
+                    self.last_run_interrupted = false;
+                    return Ok(WorkerResult::LimitReached);
+                }
+            }
         }
     }
 
@@ -1137,6 +1150,7 @@ impl<C: LlmClient> Worker<C, Mutable> {
             history: Vec::new(),
             locked_prefix_len: 0,
             turn_count: 0,
+            max_turns: None,
             turn_notifiers: Vec::new(),
             request_config: RequestConfig::default(),
             last_run_interrupted: false,
@@ -1330,6 +1344,11 @@ impl<C: LlmClient> Worker<C, Mutable> {
         self.turn_count = count;
     }
 
+    /// Set the maximum number of turns. None means unlimited.
+    pub fn set_max_turns(&mut self, max_turns: Option<u32>) {
+        self.max_turns = max_turns;
+    }
+
     /// Set the last_run_interrupted flag (for session restoration)
     pub fn set_last_run_interrupted(&mut self, interrupted: bool) {
         self.last_run_interrupted = interrupted;
@@ -1366,6 +1385,7 @@ impl<C: LlmClient> Worker<C, Mutable> {
             history: self.history,
             locked_prefix_len,
             turn_count: self.turn_count,
+            max_turns: self.max_turns,
             turn_notifiers: self.turn_notifiers,
             request_config: self.request_config,
             last_run_interrupted: self.last_run_interrupted,
@@ -1403,6 +1423,7 @@ impl<C: LlmClient> Worker<C, CacheLocked> {
             history: self.history,
             locked_prefix_len: 0,
             turn_count: self.turn_count,
+            max_turns: self.max_turns,
             turn_notifiers: self.turn_notifiers,
             request_config: self.request_config,
             last_run_interrupted: self.last_run_interrupted,

crates/manifest/src/lib.rs

@@ -2,6 +2,7 @@ mod scope;
 
 pub use scope::Scope;
 
+use std::num::NonZeroU32;
 use std::path::PathBuf;
 
 use serde::{Deserialize, Serialize};
@@ -56,6 +57,8 @@ pub struct WorkerManifest {
     #[serde(default)]
     pub max_tokens: Option<u32>,
     #[serde(default)]
+    pub max_turns: Option<NonZeroU32>,
+    #[serde(default)]
     pub temperature: Option<f32>,
 }
 
@@ -151,6 +154,55 @@ model = "llama3"
         assert!(manifest.provider.api_key_env.is_none());
     }
 
+    #[test]
+    fn parse_max_turns() {
+        let toml = r#"
+[pod]
+name = "test"
+
+[provider]
+kind = "anthropic"
+model = "claude-sonnet-4-20250514"
+
+[worker]
+max_turns = 50
+"#;
+        let manifest = PodManifest::from_toml(toml).unwrap();
+        assert_eq!(manifest.worker.max_turns.unwrap().get(), 50);
+    }
+
+    #[test]
+    fn omitted_max_turns_is_none() {
+        let toml = r#"
+[pod]
+name = "test"
+
+[provider]
+kind = "anthropic"
+model = "claude-sonnet-4-20250514"
+
+[worker]
+"#;
+        let manifest = PodManifest::from_toml(toml).unwrap();
+        assert!(manifest.worker.max_turns.is_none());
+    }
+
+    #[test]
+    fn reject_max_turns_zero() {
+        let toml = r#"
+[pod]
+name = "test"
+
+[provider]
+kind = "anthropic"
+model = "claude-sonnet-4-20250514"
+
+[worker]
+max_turns = 0
+"#;
+        assert!(PodManifest::from_toml(toml).is_err());
+    }
+
     #[test]
     fn reject_unknown_provider() {
         let toml = r#"

crates/pod/examples/pod_cli.rs

@@ -49,6 +49,7 @@ async fn main() -> Result<(), Box<dyn std::error::Error>> {
     match result {
         PodRunResult::Finished => println!("(finished)"),
         PodRunResult::Paused => println!("(paused)"),
+        PodRunResult::LimitReached => println!("(turn limit reached)"),
     }
 
     // 5. Extract the assistant's reply from history

crates/pod/src/controller.rs

@@ -11,7 +11,7 @@ use llm_worker_persistence::Store;
 use tokio::sync::{broadcast, mpsc};
 
 use crate::pod::{Pod, PodRunResult, PodError};
-use protocol::{ErrorCode, Event, Method, TurnResult};
+use protocol::{ErrorCode, Event, Method, RunResult, TurnResult};
 use crate::runtime_dir::RuntimeDir;
 use crate::shared_state::{PodSharedState, PodStatus};
 use crate::socket_server::SocketServer;
@@ -193,10 +193,15 @@ where
         tokio::select! {
             result = &mut pod_future => {
                 return match result {
-                    Ok(r) => match r {
-                        PodRunResult::Finished => PodStatus::Idle,
-                        PodRunResult::Paused => PodStatus::Paused,
-                    },
+                    Ok(r) => {
+                        let (status, run_result) = match r {
+                            PodRunResult::Finished => (PodStatus::Idle, RunResult::Finished),
+                            PodRunResult::Paused => (PodStatus::Paused, RunResult::Paused),
+                            PodRunResult::LimitReached => (PodStatus::Idle, RunResult::LimitReached),
+                        };
+                        let _ = event_tx.send(Event::RunEnd { result: run_result });
+                        status
+                    }
                     Err(e) => {
                         let code = worker_error_code(&e);
                         let _ = event_tx.send(Event::Error {

crates/pod/src/pod.rs

@@ -142,6 +142,7 @@ pub fn apply_worker_manifest<C: LlmClient>(worker: &mut Worker<C>, wm: &WorkerMa
         config.temperature = Some(temperature);
     }
     worker.set_request_config(config);
+    worker.set_max_turns(wm.max_turns.map(|n| n.get()));
 }
 
 /// Result of a Pod run.
@@ -151,6 +152,8 @@ pub enum PodRunResult {
     Finished,
     /// The LLM paused (e.g. awaiting user confirmation via a hook).
     Paused,
+    /// The worker reached its configured max_turns limit.
+    LimitReached,
 }
 
 impl From<llm_worker::WorkerResult> for PodRunResult {
@@ -158,6 +161,7 @@ impl From<llm_worker::WorkerResult> for PodRunResult {
         match r {
             llm_worker::WorkerResult::Finished => PodRunResult::Finished,
             llm_worker::WorkerResult::Paused => PodRunResult::Paused,
+            llm_worker::WorkerResult::LimitReached => PodRunResult::LimitReached,
         }
     }
 }

crates/protocol/src/lib.rs

@@ -60,6 +60,9 @@ pub enum Event {
         input_tokens: Option<u64>,
         output_tokens: Option<u64>,
     },
+    RunEnd {
+        result: RunResult,
+    },
     Error {
         code: ErrorCode,
         message: String,
@@ -83,6 +86,14 @@ pub enum TurnResult {
     Paused,
 }
 
+#[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
+#[serde(rename_all = "snake_case")]
+pub enum RunResult {
+    Finished,
+    Paused,
+    LimitReached,
+}
+
 #[derive(Debug, Clone, Copy, PartialEq, Eq, Serialize, Deserialize)]
 #[serde(rename_all = "snake_case")]
 pub enum ErrorCode {
@@ -126,6 +137,17 @@ mod tests {
         assert_eq!(parsed["data"]["text"], "Hello");
     }
 
+    #[test]
+    fn event_run_end_format() {
+        let event = Event::RunEnd {
+            result: RunResult::LimitReached,
+        };
+        let json = event.to_json_line().unwrap();
+        let parsed: serde_json::Value = serde_json::from_str(&json).unwrap();
+        assert_eq!(parsed["event"], "run_end");
+        assert_eq!(parsed["data"]["result"], "limit_reached");
+    }
+
     #[test]
     fn event_error_format() {
         let event = Event::Error {

crates/tui/src/app.rs

@@ -131,6 +131,9 @@ impl App {
                 });
                 self.scroll_to_bottom();
             }
+            Event::RunEnd { result } => {
+                self.push_status(format!("[run end] {result:?}"));
+            }
             Event::ToolCallArgsDelta { .. } => {}
         }
     }