refactor: remove Responses API, consolidate to Chat Completions (#272)

* fix: strip reasoning from LLM responses and persist assistant messages reliably

- Filter out `type: "reasoning"` output items from NEAR AI Responses API
  parsing so chain-of-thought never reaches the UI (nearai.rs)
- Rewrite clean_response with regex-based tag stripping that is
  code-aware (preserves tags inside fenced blocks and inline backticks),
  supports 9+ tag names (think, thought, reasoning, reflection, etc.),
  handles <final> extraction, pipe-delimited tags, and case/whitespace
  tolerance (reasoning.rs)
- Add Reasoning::complete() helper so all non-agentic LLM call sites
  (summarize, suggest, heartbeat, compaction) get automatic response
  cleaning; thread SafetyLayer through to those callers
- Change persist_turn from fire-and-forget tokio::spawn to awaited async
  so both user and assistant messages are written before returning,
  preventing data loss on shutdown/restart
- Pass input_count through seed_response_chain so response chaining
  delta calculation is accurate after thread hydration on restart
- Make NearAiResponse.usage optional and preserve response_id in alt
  response path for chaining continuity
- Persist session token to DB during onboarding wizard so runtime
  loads it without legacy-key fallback; suppress spurious warning on
  fresh installs
- Fix dev tool double-registration when builder already registers them
- Load dotenv/ironclaw env for doctor and status subcommands
- Reduce startup log noise (demote info→debug for skills, remove
  redundant info lines)

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* Nudge to not loop over tools continuesly

* refactor: remove Responses API, consolidate NEAR AI to Chat Completions only

The Responses API provider (nearai.rs, 1278 lines) added significant complexity
(response chaining state machine, delta message calculation, previous_response_id
persistence) for marginal benefit. This consolidates to the Chat Completions API
only, upgrading NearAiChatProvider with dual auth (session token + API key) and
401 retry for session token renewal.

- Delete src/llm/nearai.rs (Responses API provider)
- Upgrade nearai_chat.rs with SessionManager, dual auth, flexible list_models
- Remove response_id from CompletionResponse and ToolCompletionResponse
- Remove seed_response_chain/get_response_chain_id from LlmProvider trait
- Remove response chain persistence from agent (thread_ops, session)
- Remove NearAiApiMode enum and NEARAI_API_MODE config
- Clean up all wrapper providers (retry, circuit_breaker, failover, cache)
- Update documentation (CLAUDE.md, .env.example)

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat: runtime log level control via gateway UI and URL parameter

Add server-side log level switching using tracing_subscriber::reload::Layer
so the EnvFilter can be swapped at runtime without restarting. Expose via
GET/PUT /api/logs/level endpoints, a "Server: LEVEL" dropdown in the logs
toolbar, and a ?log_level=debug URL parameter for one-click activation.

Also applies cargo fmt to pre-existing files (llm/, tests/).

Co-Authored-By: Claude Opus 4.6 <[email protected]>

---------

Co-authored-by: Claude Opus 4.6 <[email protected]>

.env.example

@@ -6,21 +6,19 @@ DATABASE_POOL_SIZE=10
 # LLM_BACKEND=nearai           # default
 # Possible values: nearai, ollama, openai_compatible, openai, anthropic, tinfoil
 
-# === NEAR AI Chat (Responses API, session token auth) ===
+# === NEAR AI (Chat Completions API) ===
-# Default mode. Uses browser OAuth (GitHub/Google) on first run.
+# Two auth modes:
-# Session token stored in ~/.ironclaw/session.json automatically.
+#   1. Session token (default): Uses browser OAuth (GitHub/Google) on first run.
-# For hosting providers: set NEARAI_SESSION_TOKEN env var directly.
+#      Session token stored in ~/.ironclaw/session.json automatically.
+#      Base URL defaults to https://private.near.ai
+#   2. API key: Set NEARAI_API_KEY to use API key auth from cloud.near.ai.
+#      Base URL defaults to https://cloud-api.near.ai
 NEARAI_MODEL=zai-org/GLM-5-FP8
 NEARAI_BASE_URL=https://private.near.ai
 NEARAI_AUTH_URL=https://private.near.ai
 # NEARAI_SESSION_TOKEN=sess_...                  # hosting providers: set this
 # NEARAI_SESSION_PATH=~/.ironclaw/session.json   # optional, default shown
-
+# NEARAI_API_KEY=...                             # API key from cloud.near.ai
-# === NEAR AI Cloud (Chat Completions API, API key auth) ===
-# Auto-selected when NEARAI_API_KEY is set. Get a key from cloud.near.ai.
-# NEARAI_API_KEY=...
-# NEARAI_BASE_URL=https://cloud-api.near.ai      # default for cloud mode
-# NEARAI_API_MODE=chat_completions               # auto-detected from API key
 
 # Local LLM Providers (Ollama, LM Studio, vLLM, LiteLLM)
 

CLAUDE.md

@@ -120,8 +120,7 @@ src/
 ├── llm/                # LLM integration (multi-provider)
 │   ├── mod.rs          # Provider factory, LlmBackend enum
 │   ├── provider.rs     # LlmProvider trait, message types
-│   ├── nearai.rs       # NEAR AI Responses API provider
+│   ├── nearai_chat.rs  # NEAR AI Chat Completions provider (session token + API key auth)
-│   ├── nearai_chat.rs  # NEAR AI Chat Completions fallback
 │   ├── reasoning.rs    # Planning, tool selection, evaluation
 │   ├── session.rs      # Session token management with auto-renewal
 │   ├── circuit_breaker.rs # Circuit breaker for provider failures
@@ -339,13 +338,12 @@ LIBSQL_PATH=~/.ironclaw/ironclaw.db    # libSQL local path (default)
 # LIBSQL_AUTH_TOKEN=xxx                # Required with LIBSQL_URL
 
 # NEAR AI (when LLM_BACKEND=nearai, the default)
-# Two modes: "NEAR AI Chat" (session token) or "NEAR AI Cloud" (API key)
+# Two auth modes: session token (default) or API key
-# NEAR AI Chat (Responses API, default):
+# Session token auth (default): uses browser OAuth on first run
-NEARAI_SESSION_TOKEN=sess_...           # session token for chat-api
+NEARAI_SESSION_TOKEN=sess_...           # hosting providers: set this
 NEARAI_BASE_URL=https://private.near.ai
-# NEAR AI Cloud (Chat Completions API, auto-selected when API key is set):
+# API key auth: set NEARAI_API_KEY, base URL defaults to cloud-api.near.ai
 # NEARAI_API_KEY=...                    # API key from cloud.near.ai
-# NEARAI_BASE_URL=https://cloud-api.near.ai
 NEARAI_MODEL=claude-3-5-sonnet-20241022
 
 # Agent settings
@@ -408,11 +406,9 @@ TINFOIL_MODEL=kimi-k2-5               # Default model
 
 IronClaw supports multiple LLM backends via the `LLM_BACKEND` env var: `nearai` (default), `openai`, `anthropic`, `ollama`, `openai_compatible`, and `tinfoil`.
 
-**NEAR AI Chat** -- Uses the NEAR AI Responses API (`https://private.near.ai/v1/responses`). Authenticates with session tokens (`sess_xxx`) obtained via browser OAuth (GitHub/Google). Supports response chaining (delta-only follow-up messages) for efficient multi-turn conversations. This is the default mode when no `NEARAI_API_KEY` is set. Set `NEARAI_SESSION_TOKEN` env var for hosting providers that inject tokens via environment. Configure with `NEARAI_BASE_URL` (default: `https://private.near.ai`).
+**NEAR AI** -- Uses the Chat Completions API with dual auth support. Session token auth (default): authenticates with session tokens (`sess_xxx`) obtained via browser OAuth (GitHub/Google), base URL defaults to `https://private.near.ai`. API key auth: set `NEARAI_API_KEY` (from `cloud.near.ai`), base URL defaults to `https://cloud-api.near.ai`. Both modes use the same Chat Completions endpoint. Tool messages are flattened to plain text for compatibility. Set `NEARAI_SESSION_TOKEN` env var for hosting providers that inject tokens via environment.
 
-**NEAR AI Cloud** -- Uses the OpenAI-compatible Chat Completions API (`https://cloud-api.near.ai/v1/chat/completions`). Authenticates with API keys from `cloud.near.ai`. Auto-selected when `NEARAI_API_KEY` is set (or explicitly via `NEARAI_API_MODE=chat_completions`). Tool messages are flattened to plain text for compatibility. Configure with `NEARAI_API_KEY` and `NEARAI_BASE_URL` (default: `https://cloud-api.near.ai`).
+**Tinfoil** -- Private inference via `https://inference.tinfoil.sh/v1`. Runs models inside hardware-attested TEEs so neither Tinfoil nor the cloud provider can see prompts or responses. Uses the OpenAI-compatible Chat Completions API. Configure with `TINFOIL_API_KEY` and `TINFOIL_MODEL` (default: `kimi-k2-5`).
-
-**Tinfoil** -- Private inference via `https://inference.tinfoil.sh/v1`. Runs models inside hardware-attested TEEs so neither Tinfoil nor the cloud provider can see prompts or responses. Uses the OpenAI-compatible Chat Completions API only (not the Responses API, so tool calls are adapted to chat format). Configure with `TINFOIL_API_KEY` and `TINFOIL_MODEL` (default: `kimi-k2-5`).
 
 ## Database
 

benchmarks/src/instrumented_llm.rs

@@ -182,7 +182,6 @@ mod tests {
                 input_tokens: 100,
                 output_tokens: 50,
                 finish_reason: FinishReason::Stop,
-                response_id: None,
             })
         }
 
@@ -196,7 +195,6 @@ mod tests {
                 input_tokens: 200,
                 output_tokens: 100,
                 finish_reason: FinishReason::Stop,
-                response_id: None,
             })
         }
     }

src/agent/agent_loop.rs

@@ -358,10 +358,6 @@ impl Agent {
                         }
                     });
 
-                    tracing::info!(
-                        "Heartbeat enabled with {}s interval",
-                        hb_config.interval_secs
-                    );
                     let hygiene = self
                         .hygiene_config
                         .as_ref()
@@ -373,6 +369,7 @@ impl Agent {
                         hygiene,
                         workspace.clone(),
                         self.cheap_llm().clone(),
+                        self.safety().clone(),
                         Some(notify_tx),
                     ))
                 } else {

src/agent/commands.rs

@@ -13,7 +13,7 @@ use crate::agent::submission::SubmissionResult;
 use crate::agent::{Agent, MessageIntent};
 use crate::channels::{IncomingMessage, StatusUpdate};
 use crate::error::Error;
-use crate::llm::ChatMessage;
+use crate::llm::{ChatMessage, Reasoning};
 
 impl Agent {
     /// Handle job-related intents without turn tracking.
@@ -235,6 +235,7 @@ impl Agent {
             crate::workspace::hygiene::HygieneConfig::default(),
             workspace.clone(),
             self.llm().clone(),
+            self.safety().clone(),
         );
 
         match runner.check_heartbeat().await {
@@ -295,10 +296,11 @@ impl Agent {
             .with_max_tokens(512)
             .with_temperature(0.3);
 
-        match self.llm().complete(request).await {
+        let reasoning = Reasoning::new(self.llm().clone(), self.safety().clone());
-            Ok(response) => Ok(SubmissionResult::response(format!(
+        match reasoning.complete(request).await {
+            Ok((text, _usage)) => Ok(SubmissionResult::response(format!(
                 "Thread Summary:\n\n{}",
-                response.content.trim()
+                text.trim()
             ))),
             Err(e) => Ok(SubmissionResult::error(format!("Summarize failed: {}", e))),
         }
@@ -342,10 +344,11 @@ impl Agent {
             .with_max_tokens(512)
             .with_temperature(0.5);
 
-        match self.llm().complete(request).await {
+        let reasoning = Reasoning::new(self.llm().clone(), self.safety().clone());
-            Ok(response) => Ok(SubmissionResult::response(format!(
+        match reasoning.complete(request).await {
+            Ok((text, _usage)) => Ok(SubmissionResult::response(format!(
                 "Suggested Next Steps:\n\n{}",
-                response.content.trim()
+                text.trim()
             ))),
             Err(e) => Ok(SubmissionResult::error(format!("Suggest failed: {}", e))),
         }

src/agent/compaction.rs

@@ -12,7 +12,8 @@ use chrono::Utc;
 use crate::agent::context_monitor::{CompactionStrategy, ContextBreakdown};
 use crate::agent::session::Thread;
 use crate::error::Error;
-use crate::llm::{ChatMessage, CompletionRequest, LlmProvider};
+use crate::llm::{ChatMessage, CompletionRequest, LlmProvider, Reasoning};
+use crate::safety::SafetyLayer;
 use crate::workspace::Workspace;
 
 /// Result of a compaction operation.
@@ -33,12 +34,13 @@ pub struct CompactionResult {
 /// Compacts conversation context to stay within limits.
 pub struct ContextCompactor {
     llm: Arc<dyn LlmProvider>,
+    safety: Arc<SafetyLayer>,
 }
 
 impl ContextCompactor {
     /// Create a new context compactor.
-    pub fn new(llm: Arc<dyn LlmProvider>) -> Self {
+    pub fn new(llm: Arc<dyn LlmProvider>, safety: Arc<SafetyLayer>) -> Self {
-        Self { llm }
+        Self { llm, safety }
     }
 
     /// Compact a thread's context using the given strategy.
@@ -231,8 +233,9 @@ Be brief but capture all important details. Use bullet points."#,
             .with_max_tokens(1024)
             .with_temperature(0.3);
 
-        let response = self.llm.complete(request).await?;
+        let reasoning = Reasoning::new(self.llm.clone(), self.safety.clone());
-        Ok(response.content)
+        let (text, _) = reasoning.complete(request).await?;
+        Ok(text)
     }
 
     /// Write a summary to the workspace daily log.

src/agent/dispatcher.rs

@@ -109,10 +109,17 @@ impl Agent {
         let job_ctx = JobContext::with_user(&message.user_id, "chat", "Interactive chat session");
 
         const MAX_TOOL_ITERATIONS: usize = 10;
+        // Force a text-only response on the last iteration to guarantee termination
+        // instead of hard-erroring. The penultimate iteration also gets a nudge
+        // message so the LLM knows it should wrap up.
+        const FORCE_TEXT_AT: usize = MAX_TOOL_ITERATIONS;
+        const NUDGE_AT: usize = MAX_TOOL_ITERATIONS - 1;
         let mut iteration = 0;
         loop {
             iteration += 1;
-            if iteration > MAX_TOOL_ITERATIONS {
+            // Hard ceiling one past the forced-text iteration (should never be reached
+            // since FORCE_TEXT_AT guarantees a text response, but kept as a safety net).
+            if iteration > MAX_TOOL_ITERATIONS + 1 {
                 return Err(crate::error::LlmError::InvalidResponse {
                     provider: "agent".to_string(),
                     reason: format!("Exceeded maximum tool iterations ({})", MAX_TOOL_ITERATIONS),
@@ -143,6 +150,19 @@ impl Agent {
                 .into());
             }
 
+            // Inject a nudge message when approaching the iteration limit so the
+            // LLM is aware it should produce a final answer on the next turn.
+            if iteration == NUDGE_AT {
+                context_messages.push(ChatMessage::system(
+                    "You are approaching the tool call limit. \
+                     Provide your best final answer on the next response \
+                     using the information you have gathered so far. \
+                     Do not call any more tools.",
+                ));
+            }
+
+            let force_text = iteration >= FORCE_TEXT_AT;
+
             // Refresh tool definitions each iteration so newly built tools become visible
             let tool_defs = self.tools().tool_definitions().await;
 
@@ -162,8 +182,9 @@ impl Agent {
                 tool_defs
             };
 
-            // Call LLM with current context
+            // Call LLM with current context; force_text drops tools to guarantee a
-            let context = ReasoningContext::new()
+            // text response on the final iteration.
+            let mut context = ReasoningContext::new()
                 .with_messages(context_messages.clone())
                 .with_tools(tool_defs)
                 .with_metadata({
@@ -171,6 +192,14 @@ impl Agent {
                     m.insert("thread_id".to_string(), thread_id.to_string());
                     m
                 });
+            context.force_text = force_text;
+
+            if force_text {
+                tracing::info!(
+                    iteration,
+                    "Forcing text-only response (iteration limit reached)"
+                );
+            }
 
             let output = reasoning.respond_with_tools(&context).await?;
 
@@ -799,7 +828,6 @@ mod tests {
                 input_tokens: 0,
                 output_tokens: 0,
                 finish_reason: FinishReason::Stop,
-                response_id: None,
             })
         }
 
@@ -813,7 +841,6 @@ mod tests {
                 input_tokens: 0,
                 output_tokens: 0,
                 finish_reason: FinishReason::Stop,
-                response_id: None,
             })
         }
     }

src/agent/heartbeat.rs

@@ -29,7 +29,8 @@ use std::time::Duration;
 use tokio::sync::mpsc;
 
 use crate::channels::OutgoingResponse;
-use crate::llm::{ChatMessage, CompletionRequest, FinishReason, LlmProvider};
+use crate::llm::{ChatMessage, CompletionRequest, LlmProvider, Reasoning};
+use crate::safety::SafetyLayer;
 use crate::workspace::Workspace;
 use crate::workspace::hygiene::HygieneConfig;
 
@@ -100,6 +101,7 @@ pub struct HeartbeatRunner {
     hygiene_config: HygieneConfig,
     workspace: Arc<Workspace>,
     llm: Arc<dyn LlmProvider>,
+    safety: Arc<SafetyLayer>,
     response_tx: Option<mpsc::Sender<OutgoingResponse>>,
     consecutive_failures: u32,
 }
@@ -111,12 +113,14 @@ impl HeartbeatRunner {
         hygiene_config: HygieneConfig,
         workspace: Arc<Workspace>,
         llm: Arc<dyn LlmProvider>,
+        safety: Arc<SafetyLayer>,
     ) -> Self {
         Self {
             config,
             hygiene_config,
             workspace,
             llm,
+            safety,
             response_tx: None,
             consecutive_failures: 0,
         }
@@ -258,25 +262,18 @@ impl HeartbeatRunner {
             .with_max_tokens(max_tokens)
             .with_temperature(0.3);
 
-        let response = match self.llm.complete(request).await {
+        let reasoning = Reasoning::new(self.llm.clone(), self.safety.clone());
+        let (content, _usage) = match reasoning.complete(request).await {
             Ok(r) => r,
             Err(e) => return HeartbeatResult::Failed(format!("LLM call failed: {}", e)),
         };
 
-        let content = response.content.trim();
+        let content = content.trim();
 
         // Guard against empty content. Reasoning models (e.g. GLM-4.7) may
         // burn all output tokens on chain-of-thought and return content: null.
         if content.is_empty() {
-            return if response.finish_reason == FinishReason::Length {
+            return HeartbeatResult::Failed("LLM returned empty content.".to_string());
-                HeartbeatResult::Failed(
-                    "LLM response was truncated (finish_reason=length) with no content. \
-                     The model may have exhausted its token budget on reasoning."
-                        .to_string(),
-                )
-            } else {
-                HeartbeatResult::Failed("LLM returned empty content.".to_string())
-            };
         }
 
         // Check if nothing needs attention
@@ -355,9 +352,10 @@ pub fn spawn_heartbeat(
     hygiene_config: HygieneConfig,
     workspace: Arc<Workspace>,
     llm: Arc<dyn LlmProvider>,
+    safety: Arc<SafetyLayer>,
     response_tx: Option<mpsc::Sender<OutgoingResponse>>,
 ) -> tokio::task::JoinHandle<()> {
-    let mut runner = HeartbeatRunner::new(config, hygiene_config, workspace, llm);
+    let mut runner = HeartbeatRunner::new(config, hygiene_config, workspace, llm, safety);
     if let Some(tx) = response_tx {
         runner = runner.with_response_channel(tx);
     }

src/agent/session.rs

@@ -185,10 +185,6 @@ pub struct Thread {
     /// Pending auth token request (thread is in auth mode).
     #[serde(default)]
     pub pending_auth: Option<PendingAuth>,
-    /// Last NEAR AI response ID for response chaining. Persisted to DB
-    /// metadata so we can resume chaining across restarts.
-    #[serde(default)]
-    pub last_response_id: Option<String>,
 }
 
 impl Thread {
@@ -205,7 +201,6 @@ impl Thread {
             metadata: serde_json::Value::Null,
             pending_approval: None,
             pending_auth: None,
-            last_response_id: None,
         }
     }
 
@@ -222,7 +217,6 @@ impl Thread {
             metadata: serde_json::Value::Null,
             pending_approval: None,
             pending_auth: None,
-            last_response_id: None,
         }
     }
 
@@ -863,7 +857,6 @@ mod tests {
 
         thread.start_turn("hello");
         thread.complete_turn("world");
-        thread.last_response_id = Some("resp_abc123".to_string());
 
         let json = serde_json::to_string(&thread).unwrap();
         let restored: Thread = serde_json::from_str(&json).unwrap();
@@ -873,7 +866,6 @@ mod tests {
         assert_eq!(restored.turns.len(), 1);
         assert_eq!(restored.turns[0].user_input, "hello");
         assert_eq!(restored.turns[0].response, Some("world".to_string()));
-        assert_eq!(restored.last_response_id, Some("resp_abc123".to_string()));
     }
 
     #[test]

src/agent/thread_ops.rs

@@ -87,20 +87,6 @@ impl Agent {
             thread.restore_from_messages(chat_messages);
         }
 
-        // Restore response chain from conversation metadata
-        if let Some(store) = self.store()
-            && let Ok(Some(metadata)) = store.get_conversation_metadata(thread_uuid).await
-            && let Some(rid) = metadata
-                .get("last_response_id")
-                .and_then(|v| v.as_str())
-                .map(String::from)
-        {
-            thread.last_response_id = Some(rid.clone());
-            self.llm()
-                .seed_response_chain(&thread_uuid.to_string(), rid);
-            tracing::debug!("Restored response chain for thread {}", thread_uuid);
-        }
-
         // Insert into session and register with session manager
         {
             let mut sess = session.lock().await;
@@ -228,7 +214,7 @@ impl Agent {
                     )
                     .await;
 
-                let compactor = ContextCompactor::new(self.llm().clone());
+                let compactor = ContextCompactor::new(self.llm().clone(), self.safety().clone());
                 if let Err(e) = compactor
                     .compact(thread, strategy, self.workspace().map(|w| w.as_ref()))
                     .await
@@ -325,7 +311,6 @@ impl Agent {
                 };
 
                 thread.complete_turn(&response);
-                self.persist_response_chain(thread);
                 let _ = self
                     .channels
                     .send_status(
@@ -335,8 +320,10 @@ impl Agent {
                     )
                     .await;
 
-                // Fire-and-forget: persist turn to DB
+                // Persist turn to DB before returning so the write
-                self.persist_turn(thread_id, &message.user_id, content, Some(&response));
+                // completes even if the process shuts down right after.
+                self.persist_turn(thread_id, &message.user_id, content, Some(&response))
+                    .await;
 
                 Ok(SubmissionResult::response(response))
             }
@@ -366,15 +353,16 @@ impl Agent {
                 thread.fail_turn(e.to_string());
 
                 // Persist the user message even on failure
-                self.persist_turn(thread_id, &message.user_id, content, None);
+                self.persist_turn(thread_id, &message.user_id, content, None)
+                    .await;
 
                 Ok(SubmissionResult::error(e.to_string()))
             }
         }
     }
 
-    /// Fire-and-forget: persist a turn (user message + optional assistant response) to the DB.
+    /// Persist a turn (user message + optional assistant response) to the DB.
-    pub(super) fn persist_turn(
+    pub(super) async fn persist_turn(
         &self,
         thread_id: Uuid,
         user_id: &str,
@@ -386,70 +374,29 @@ impl Agent {
             None => return,
         };
 
-        let user_id = user_id.to_string();
+        if let Err(e) = store
-        let user_input = user_input.to_string();
+            .ensure_conversation(thread_id, "gateway", user_id, None)
-        let response = response.map(String::from);
+            .await
+        {
+            tracing::warn!("Failed to ensure conversation {}: {}", thread_id, e);
+            return;
+        }
 
-        tokio::spawn(async move {
+        if let Err(e) = store
-            if let Err(e) = store
+            .add_conversation_message(thread_id, "user", user_input)
-                .ensure_conversation(thread_id, "gateway", &user_id, None)
+            .await
+        {
+            tracing::warn!("Failed to persist user message: {}", e);
+            return;
+        }
+
+        if let Some(resp) = response
+            && let Err(e) = store
+                .add_conversation_message(thread_id, "assistant", resp)
                 .await
-            {
+        {
-                tracing::warn!("Failed to ensure conversation {}: {}", thread_id, e);
+            tracing::warn!("Failed to persist assistant message: {}", e);
-                return;
+        }
-            }
-
-            if let Err(e) = store
-                .add_conversation_message(thread_id, "user", &user_input)
-                .await
-            {
-                tracing::warn!("Failed to persist user message: {}", e);
-                return;
-            }
-
-            if let Some(ref resp) = response
-                && let Err(e) = store
-                    .add_conversation_message(thread_id, "assistant", resp)
-                    .await
-            {
-                tracing::warn!("Failed to persist assistant message: {}", e);
-            }
-        });
-    }
-
-    /// Sync the provider's response chain ID to the thread and DB metadata.
-    ///
-    /// Call after a successful agentic loop to persist the latest
-    /// `previous_response_id` so chaining survives restarts.
-    pub(super) fn persist_response_chain(&self, thread: &mut crate::agent::session::Thread) {
-        let tid = thread.id.to_string();
-        let response_id = match self.llm().get_response_chain_id(&tid) {
-            Some(rid) => rid,
-            None => return,
-        };
-
-        // Update in-memory thread
-        thread.last_response_id = Some(response_id.clone());
-
-        // Fire-and-forget DB write
-        let store = match self.store() {
-            Some(s) => Arc::clone(s),
-            None => return,
-        };
-        let thread_id = thread.id;
-        tokio::spawn(async move {
-            let val = serde_json::json!(response_id);
-            if let Err(e) = store
-                .update_conversation_metadata_field(thread_id, "last_response_id", &val)
-                .await
-            {
-                tracing::warn!(
-                    "Failed to persist response chain for thread {}: {}",
-                    thread_id,
-                    e
-                );
-            }
-        });
     }
 
     pub(super) async fn process_undo(
@@ -562,7 +509,7 @@ impl Agent {
                 crate::agent::context_monitor::CompactionStrategy::Summarize { keep_recent: 5 },
             );
 
-        let compactor = ContextCompactor::new(self.llm().clone());
+        let compactor = ContextCompactor::new(self.llm().clone(), self.safety().clone());
         match compactor
             .compact(thread, strategy, self.workspace().map(|w| w.as_ref()))
             .await
@@ -1072,9 +1019,9 @@ impl Agent {
                     let user_input = thread.last_turn().map(|t| t.user_input.clone());
                     thread.complete_turn(&response);
                     if let Some(input) = user_input {
-                        self.persist_turn(thread_id, &message.user_id, &input, Some(&response));
+                        self.persist_turn(thread_id, &message.user_id, &input, Some(&response))
+                            .await;
                     }
-                    self.persist_response_chain(thread);
                     let _ = self
                         .channels
                         .send_status(
@@ -1112,7 +1059,8 @@ impl Agent {
                     let user_input = thread.last_turn().map(|t| t.user_input.clone());
                     thread.fail_turn(e.to_string());
                     if let Some(input) = user_input {
-                        self.persist_turn(thread_id, &message.user_id, &input, None);
+                        self.persist_turn(thread_id, &message.user_id, &input, None)
+                            .await;
                     }
                     Ok(SubmissionResult::error(e.to_string()))
                 }
@@ -1131,7 +1079,8 @@ impl Agent {
                     thread.clear_pending_approval();
                     thread.complete_turn(&rejection);
                     if let Some(input) = user_input {
-                        self.persist_turn(thread_id, &message.user_id, &input, Some(&rejection));
+                        self.persist_turn(thread_id, &message.user_id, &input, Some(&rejection))
+                            .await;
                     }
                 }
             }
@@ -1171,9 +1120,9 @@ impl Agent {
                 thread.enter_auth_mode(ext_name.clone());
                 thread.complete_turn(&instructions);
                 if let Some(input) = user_input {
-                    self.persist_turn(thread_id, &message.user_id, &input, Some(&instructions));
+                    self.persist_turn(thread_id, &message.user_id, &input, Some(&instructions))
+                        .await;
                 }
-                self.persist_response_chain(thread);
             }
         }
         let _ = self

src/app.rs

@@ -396,7 +396,6 @@ impl AppBuilder {
 
         let tools = Arc::new(ToolRegistry::new());
         tools.register_builtin_tools();
-        tracing::info!("Registered {} built-in tools", tools.count());
 
         // Create embeddings provider if configured
         let embeddings: Option<Arc<dyn EmbeddingProvider>> = if self.config.embeddings.enabled {
@@ -681,12 +680,12 @@ impl AppBuilder {
             None
         };
 
-        // Register dev tools if local tools are enabled
+        // register_builder_tool() already calls register_dev_tools() internally,
-        if self.config.agent.allow_local_tools {
+        // so only register them here when the builder didn't already do it.
+        let builder_registered_dev_tools = self.config.builder.enabled
+            && (self.config.agent.allow_local_tools || !self.config.sandbox.enabled);
+        if self.config.agent.allow_local_tools && !builder_registered_dev_tools {
             tools.register_dev_tools();
-            tracing::info!(
-                "Local tools enabled (allow_local_tools=true), dev tools registered directly"
-            );
         }
 
         Ok((mcp_session_manager, wasm_tool_runtime, extension_manager))
@@ -709,9 +708,6 @@ impl AppBuilder {
         // Seed workspace and backfill embeddings
         if let Some(ref ws) = workspace {
             match ws.seed_if_empty().await {
-                Ok(count) if count > 0 => {
-                    tracing::info!("Workspace seeded with {} core files", count);
-                }
                 Ok(_) => {}
                 Err(e) => {
                     tracing::warn!("Failed to seed workspace: {}", e);

src/channels/web/log_layer.rs

@@ -22,7 +22,9 @@ use std::sync::{Arc, Mutex};
 use serde::Serialize;
 use tokio::sync::broadcast;
 use tracing::field::{Field, Visit};
-use tracing_subscriber::Layer;
+use tracing_subscriber::layer::SubscriberExt;
+use tracing_subscriber::util::SubscriberInitExt;
+use tracing_subscriber::{EnvFilter, Layer, reload};
 
 use crate::safety::LeakDetector;
 
@@ -102,6 +104,115 @@ impl Default for LogBroadcaster {
     }
 }
 
+/// Handle for changing the tracing `EnvFilter` at runtime.
+///
+/// Wraps a `reload::Handle` so the gateway can switch between log levels
+/// (e.g. `ironclaw=debug`) without restarting the process.
+pub struct LogLevelHandle {
+    handle: reload::Handle<EnvFilter, tracing_subscriber::Registry>,
+    current_level: Mutex<String>,
+    base_filter: String,
+}
+
+impl LogLevelHandle {
+    pub fn new(
+        handle: reload::Handle<EnvFilter, tracing_subscriber::Registry>,
+        initial_level: String,
+        base_filter: String,
+    ) -> Self {
+        Self {
+            handle,
+            current_level: Mutex::new(initial_level),
+            base_filter,
+        }
+    }
+
+    /// Change the `ironclaw=<level>` directive at runtime.
+    ///
+    /// `level` must be one of: trace, debug, info, warn, error.
+    pub fn set_level(&self, level: &str) -> Result<(), String> {
+        const VALID: &[&str] = &["trace", "debug", "info", "warn", "error"];
+        let level = level.to_lowercase();
+        if !VALID.contains(&level.as_str()) {
+            return Err(format!(
+                "invalid level '{}', must be one of: {}",
+                level,
+                VALID.join(", ")
+            ));
+        }
+
+        let filter_str = if self.base_filter.is_empty() {
+            format!("ironclaw={}", level)
+        } else {
+            format!("ironclaw={},{}", level, self.base_filter)
+        };
+
+        let new_filter = EnvFilter::new(&filter_str);
+        self.handle
+            .reload(new_filter)
+            .map_err(|e| format!("failed to reload filter: {}", e))?;
+
+        if let Ok(mut current) = self.current_level.lock() {
+            *current = level;
+        }
+        Ok(())
+    }
+
+    /// Returns the current ironclaw log level (e.g. "info", "debug").
+    pub fn current_level(&self) -> String {
+        self.current_level
+            .lock()
+            .map(|l| l.clone())
+            .unwrap_or_else(|_| "info".to_string())
+    }
+}
+
+/// Initialise the tracing subscriber with a reloadable `EnvFilter`.
+///
+/// Returns the `LogLevelHandle` so callers can swap the filter at runtime.
+/// The fmt layer and `WebLogLayer` are attached alongside the reloadable filter.
+pub fn init_tracing(log_broadcaster: Arc<LogBroadcaster>) -> Arc<LogLevelHandle> {
+    let raw_filter =
+        std::env::var("RUST_LOG").unwrap_or_else(|_| "ironclaw=info,tower_http=warn".to_string());
+
+    // Split into the ironclaw directive and "everything else" (base_filter).
+    let mut ironclaw_level = String::from("info");
+    let mut base_parts: Vec<&str> = Vec::new();
+
+    for part in raw_filter.split(',') {
+        let trimmed = part.trim();
+        if trimmed.starts_with("ironclaw=") {
+            if let Some(lvl) = trimmed.strip_prefix("ironclaw=") {
+                ironclaw_level = lvl.to_string();
+            }
+        } else if !trimmed.is_empty() {
+            base_parts.push(trimmed);
+        }
+    }
+    let base_filter = base_parts.join(",");
+
+    let env_filter = EnvFilter::new(&raw_filter);
+    let (reload_layer, reload_handle) = reload::Layer::new(env_filter);
+
+    let handle = Arc::new(LogLevelHandle::new(
+        reload_handle,
+        ironclaw_level,
+        base_filter,
+    ));
+
+    tracing_subscriber::registry()
+        .with(reload_layer)
+        .with(
+            tracing_subscriber::fmt::layer()
+                .with_target(false)
+                .with_writer(crate::tracing_fmt::TruncatingStderr::default()),
+        )
+        .with(WebLogLayer::new(log_broadcaster))
+        .init();
+
+    handle
+}
+
 /// Visitor that extracts the `message` field and all extra key-value
 /// fields from a tracing event.
 ///

src/channels/web/mod.rs

@@ -41,7 +41,7 @@ use crate::skills::registry::SkillRegistry;
 use crate::tools::ToolRegistry;
 use crate::workspace::Workspace;
 
-use self::log_layer::LogBroadcaster;
+use self::log_layer::{LogBroadcaster, LogLevelHandle};
 
 use self::server::GatewayState;
 use self::sse::SseManager;
@@ -76,6 +76,7 @@ impl GatewayChannel {
             workspace: None,
             session_manager: None,
             log_broadcaster: None,
+            log_level_handle: None,
             extension_manager: None,
             tool_registry: None,
             store: None,
@@ -105,6 +106,7 @@ impl GatewayChannel {
             workspace: self.state.workspace.clone(),
             session_manager: self.state.session_manager.clone(),
             log_broadcaster: self.state.log_broadcaster.clone(),
+            log_level_handle: self.state.log_level_handle.clone(),
             extension_manager: self.state.extension_manager.clone(),
             tool_registry: self.state.tool_registry.clone(),
             store: self.state.store.clone(),
@@ -140,6 +142,12 @@ impl GatewayChannel {
         self
     }
 
+    /// Inject the log level handle for runtime log level control.
+    pub fn with_log_level_handle(mut self, h: Arc<LogLevelHandle>) -> Self {
+        self.rebuild_state(|s| s.log_level_handle = Some(h));
+        self
+    }
+
     /// Inject the extension manager for the extensions API.
     pub fn with_extension_manager(mut self, em: Arc<ExtensionManager>) -> Self {
         self.rebuild_state(|s| s.extension_manager = Some(em));

src/channels/web/server.rs

@@ -122,6 +122,8 @@ pub struct GatewayState {
     pub session_manager: Option<Arc<SessionManager>>,
     /// Log broadcaster for the logs SSE endpoint.
     pub log_broadcaster: Option<Arc<LogBroadcaster>>,
+    /// Handle for changing the tracing log level at runtime.
+    pub log_level_handle: Option<Arc<crate::channels::web::log_layer::LogLevelHandle>>,
     /// Extension manager for extension management API.
     pub extension_manager: Option<Arc<ExtensionManager>>,
     /// Tool registry for listing registered tools.
@@ -204,6 +206,11 @@ pub async fn start_server(
         .route("/api/jobs/{id}/files/read", get(job_files_read_handler))
         // Logs
         .route("/api/logs/events", get(logs_events_handler))
+        .route("/api/logs/level", get(logs_level_get_handler))
+        .route(
+            "/api/logs/level",
+            axum::routing::put(logs_level_set_handler),
+        )
         // Extensions
         .route("/api/extensions", get(extensions_list_handler))
         .route("/api/extensions/tools", get(extensions_tools_handler))
@@ -1620,6 +1627,38 @@ async fn logs_events_handler(
     ))
 }
 
+async fn logs_level_get_handler(
+    State(state): State<Arc<GatewayState>>,
+) -> Result<Json<serde_json::Value>, (StatusCode, String)> {
+    let handle = state.log_level_handle.as_ref().ok_or((
+        StatusCode::SERVICE_UNAVAILABLE,
+        "Log level control not available".to_string(),
+    ))?;
+    Ok(Json(serde_json::json!({ "level": handle.current_level() })))
+}
+
+async fn logs_level_set_handler(
+    State(state): State<Arc<GatewayState>>,
+    Json(body): Json<serde_json::Value>,
+) -> Result<Json<serde_json::Value>, (StatusCode, String)> {
+    let handle = state.log_level_handle.as_ref().ok_or((
+        StatusCode::SERVICE_UNAVAILABLE,
+        "Log level control not available".to_string(),
+    ))?;
+
+    let level = body
+        .get("level")
+        .and_then(|v| v.as_str())
+        .ok_or((StatusCode::BAD_REQUEST, "missing 'level' field".to_string()))?;
+
+    handle
+        .set_level(level)
+        .map_err(|e| (StatusCode::BAD_REQUEST, e))?;
+
+    tracing::info!("Log level changed to '{}'", handle.current_level());
+    Ok(Json(serde_json::json!({ "level": handle.current_level() })))
+}
+
 // --- Extension handlers ---
 
 async fn extensions_list_handler(

src/channels/web/static/app.js

@@ -29,9 +29,11 @@ function authenticate() {
       sessionStorage.setItem('ironclaw_token', token);
       document.getElementById('auth-screen').style.display = 'none';
       document.getElementById('app').style.display = 'flex';
-      // Strip token from URL so it's not visible in the address bar
+      // Strip token and log_level from URL so they're not visible in the address bar
       const cleaned = new URL(window.location);
+      const urlLogLevel = cleaned.searchParams.get('log_level');
       cleaned.searchParams.delete('token');
+      cleaned.searchParams.delete('log_level');
       window.history.replaceState({}, '', cleaned.pathname + cleaned.search);
       connectSSE();
       connectLogSSE();
@@ -39,6 +41,12 @@ function authenticate() {
       loadThreads();
       loadMemoryTree();
       loadJobs();
+      // Apply URL log_level param if present, otherwise just sync the dropdown
+      if (urlLogLevel) {
+        setServerLogLevel(urlLogLevel);
+      } else {
+        loadServerLogLevel();
+      }
     })
     .catch(() => {
       sessionStorage.removeItem('ironclaw_token');
@@ -1167,6 +1175,30 @@ function applyLogFilters() {
   }
 }
 
+// --- Server-side log level control ---
+
+function setServerLogLevel(level) {
+  apiFetch('/api/logs/level', {
+    method: 'PUT',
+    headers: { 'Content-Type': 'application/json' },
+    body: JSON.stringify({ level: level }),
+  })
+    .then(r => r.json())
+    .then(data => {
+      document.getElementById('logs-server-level').value = data.level;
+    })
+    .catch(err => console.error('Failed to set server log level:', err));
+}
+
+function loadServerLogLevel() {
+  apiFetch('/api/logs/level')
+    .then(r => r.json())
+    .then(data => {
+      document.getElementById('logs-server-level').value = data.level;
+    })
+    .catch(() => {}); // ignore if not available
+}
+
 // --- Extensions ---
 
 function loadExtensions() {

src/channels/web/static/index.html

@@ -127,6 +127,12 @@
     <div class="tab-panel" id="tab-logs">
       <div class="logs-container">
         <div class="logs-toolbar">
+          <select id="logs-server-level" onchange="setServerLogLevel(this.value)" title="Server-side log level (changes what the server emits)">
+            <option value="error">Server: ERROR</option>
+            <option value="warn">Server: WARN</option>
+            <option value="info" selected>Server: INFO</option>
+            <option value="debug">Server: DEBUG</option>
+          </select>
           <select id="logs-level-filter">
             <option value="all">All Levels</option>
             <option value="ERROR">Error</option>

src/channels/web/ws.rs

@@ -477,6 +477,7 @@ mod tests {
             workspace: None,
             session_manager: None,
             log_broadcaster: None,
+            log_level_handle: None,
             extension_manager: None,
             tool_registry: None,
             store: None,

src/config/llm.rs

@@ -121,37 +121,7 @@ pub struct LlmConfig {
     pub tinfoil: Option<TinfoilConfig>,
 }
 
-/// API mode for NEAR AI.
+/// NEAR AI configuration.
-///
-/// - `Responses` = **NEAR AI Chat** (`private.near.ai`, session token auth)
-/// - `ChatCompletions` = **NEAR AI Cloud** (`cloud-api.near.ai`, API key auth)
-#[derive(Debug, Clone, Copy, PartialEq, Eq, Default)]
-pub enum NearAiApiMode {
-    /// NEAR AI Chat: Responses API with session token auth
-    #[default]
-    Responses,
-    /// NEAR AI Cloud: Chat Completions API with API key auth
-    ChatCompletions,
-}
-
-impl std::str::FromStr for NearAiApiMode {
-    type Err = String;
-
-    fn from_str(s: &str) -> Result<Self, Self::Err> {
-        match s.to_lowercase().as_str() {
-            "responses" | "response" => Ok(Self::Responses),
-            "chat_completions" | "chatcompletions" | "chat" | "completions" => {
-                Ok(Self::ChatCompletions)
-            }
-            _ => Err(format!(
-                "invalid API mode '{}', expected 'responses' or 'chat_completions'",
-                s
-            )),
-        }
-    }
-}
-
-/// NEAR AI configuration (shared by Chat and Cloud modes).
 #[derive(Debug, Clone)]
 pub struct NearAiConfig {
     /// Model to use (e.g., "claude-3-5-sonnet-20241022", "gpt-4o")
@@ -160,16 +130,13 @@ pub struct NearAiConfig {
     /// Falls back to the main model if not set.
     pub cheap_model: Option<String>,
     /// Base URL for the NEAR AI API.
-    /// Chat mode default: `https://private.near.ai`
+    /// Default: `https://private.near.ai` (session token) or `https://cloud-api.near.ai` (API key)
-    /// Cloud mode default: `https://cloud-api.near.ai`
     pub base_url: String,
     /// Base URL for auth/refresh endpoints (default: https://private.near.ai)
     pub auth_base_url: String,
     /// Path to session file (default: ~/.ironclaw/session.json)
     pub session_path: PathBuf,
-    /// API mode: NEAR AI Chat (Responses) or NEAR AI Cloud (ChatCompletions)
+    /// API key for NEAR AI Cloud. When set, uses API key auth; otherwise uses session token auth.
-    pub api_mode: NearAiApiMode,
-    /// API key for NEAR AI Cloud (required for ChatCompletions mode)
     pub api_key: Option<SecretString>,
     /// Optional fallback model for failover (default: None).
     /// When set, a secondary provider is created with this model and wrapped
@@ -229,17 +196,6 @@ impl LlmConfig {
         // Resolve NEAR AI config only when backend is NearAi (or when explicitly configured)
         let nearai_api_key = optional_env("NEARAI_API_KEY")?.map(SecretString::from);
 
-        let api_mode = if let Some(mode_str) = optional_env("NEARAI_API_MODE")? {
-            mode_str.parse().map_err(|e| ConfigError::InvalidValue {
-                key: "NEARAI_API_MODE".to_string(),
-                message: e,
-            })?
-        } else if nearai_api_key.is_some() {
-            NearAiApiMode::ChatCompletions
-        } else {
-            NearAiApiMode::Responses
-        };
-
         let nearai = NearAiConfig {
             model: optional_env("NEARAI_MODEL")?
                 .or_else(|| settings.selected_model.clone())
@@ -249,7 +205,7 @@ impl LlmConfig {
                 }),
             cheap_model: optional_env("NEARAI_CHEAP_MODEL")?,
             base_url: optional_env("NEARAI_BASE_URL")?.unwrap_or_else(|| {
-                if api_mode == NearAiApiMode::ChatCompletions {
+                if nearai_api_key.is_some() {
                     "https://cloud-api.near.ai".to_string()
                 } else {
                     "https://private.near.ai".to_string()
@@ -260,7 +216,6 @@ impl LlmConfig {
             session_path: optional_env("NEARAI_SESSION_PATH")?
                 .map(PathBuf::from)
                 .unwrap_or_else(default_session_path),
-            api_mode,
             api_key: nearai_api_key,
             fallback_model: optional_env("NEARAI_FALLBACK_MODEL")?,
             max_retries: parse_optional_env("NEARAI_MAX_RETRIES", 3)?,

src/config/mod.rs

@@ -37,7 +37,7 @@ pub use self::embeddings::EmbeddingsConfig;
 pub use self::heartbeat::HeartbeatConfig;
 pub use self::hygiene::HygieneConfig;
 pub use self::llm::{
-    AnthropicDirectConfig, LlmBackend, LlmConfig, NearAiApiMode, NearAiConfig, OllamaConfig,
+    AnthropicDirectConfig, LlmBackend, LlmConfig, NearAiConfig, OllamaConfig,
     OpenAiCompatibleConfig, OpenAiDirectConfig, TinfoilConfig,
 };
 pub use self::routines::RoutineConfig;

src/llm/circuit_breaker.rs

@@ -296,14 +296,6 @@ impl LlmProvider for CircuitBreakerProvider {
         self.inner.set_model(model)
     }
 
-    fn seed_response_chain(&self, thread_id: &str, response_id: String) {
-        self.inner.seed_response_chain(thread_id, response_id)
-    }
-
-    fn get_response_chain_id(&self, thread_id: &str) -> Option<String> {
-        self.inner.get_response_chain_id(thread_id)
-    }
-
     fn calculate_cost(&self, input_tokens: u32, output_tokens: u32) -> Decimal {
         self.inner.calculate_cost(input_tokens, output_tokens)
     }

src/llm/failover.rs

@@ -359,15 +359,6 @@ impl LlmProvider for FailoverProvider {
             .await
     }
 
-    fn seed_response_chain(&self, thread_id: &str, response_id: String) {
-        self.providers[self.last_used.load(Ordering::Relaxed)]
-            .seed_response_chain(thread_id, response_id);
-    }
-
-    fn get_response_chain_id(&self, thread_id: &str) -> Option<String> {
-        self.providers[self.last_used.load(Ordering::Relaxed)].get_response_chain_id(thread_id)
-    }
-
     fn calculate_cost(&self, input_tokens: u32, output_tokens: u32) -> Decimal {
         self.providers[self.last_used.load(Ordering::Relaxed)]
             .calculate_cost(input_tokens, output_tokens)
@@ -413,7 +404,6 @@ mod tests {
                     input_tokens: 10,
                     output_tokens: 5,
                     finish_reason: FinishReason::Stop,
-                    response_id: None,
                 }))),
                 tool_complete_result: Mutex::new(Some(Ok(ToolCompletionResponse {
                     content: Some(content.to_string()),
@@ -421,7 +411,6 @@ mod tests {
                     input_tokens: 10,
                     output_tokens: 5,
                     finish_reason: FinishReason::Stop,
-                    response_id: None,
                 }))),
             }
         }
@@ -803,7 +792,6 @@ mod tests {
                 input_tokens: 10,
                 output_tokens: 5,
                 finish_reason: FinishReason::Stop,
-                response_id: None,
             })
         }
 
@@ -829,7 +817,6 @@ mod tests {
                 input_tokens: 10,
                 output_tokens: 5,
                 finish_reason: FinishReason::Stop,
-                response_id: None,
             })
         }
 

src/llm/mod.rs

@@ -1,7 +1,7 @@
 //! LLM integration for the agent.
 //!
 //! Supports multiple backends:
-//! - **NEAR AI** (default): Session-based or API key auth via NEAR AI proxy
+//! - **NEAR AI** (default): Session token or API key auth via Chat Completions API
 //! - **OpenAI**: Direct API access with your own key
 //! - **Anthropic**: Direct API access with your own key
 //! - **Ollama**: Local model inference
@@ -10,7 +10,6 @@
 pub mod circuit_breaker;
 pub mod costs;
 pub mod failover;
-mod nearai;
 mod nearai_chat;
 mod provider;
 mod reasoning;
@@ -21,8 +20,7 @@ pub mod session;
 
 pub use circuit_breaker::{CircuitBreakerConfig, CircuitBreakerProvider};
 pub use failover::{CooldownConfig, FailoverProvider};
-pub use nearai::{ModelInfo, NearAiProvider};
+pub use nearai_chat::{ModelInfo, NearAiChatProvider};
-pub use nearai_chat::NearAiChatProvider;
 pub use provider::{
     ChatMessage, CompletionRequest, CompletionResponse, FinishReason, LlmProvider, ModelMetadata,
     Role, ToolCall, ToolCompletionRequest, ToolCompletionResponse, ToolDefinition, ToolResult,
@@ -41,7 +39,7 @@ use std::sync::Arc;
 use rig::client::CompletionClient;
 use secrecy::ExposeSecret;
 
-use crate::config::{LlmBackend, LlmConfig, NearAiApiMode, NearAiConfig};
+use crate::config::{LlmBackend, LlmConfig, NearAiConfig};
 use crate::error::LlmError;
 
 /// Create an LLM provider based on configuration.
@@ -71,24 +69,18 @@ pub fn create_llm_provider_with_config(
     config: &NearAiConfig,
     session: Arc<SessionManager>,
 ) -> Result<Arc<dyn LlmProvider>, LlmError> {
-    match config.api_mode {
+    let auth_mode = if config.api_key.is_some() {
-        NearAiApiMode::Responses => {
+        "API key"
-            tracing::info!(
+    } else {
-                model = %config.model,
+        "session token"
-                base_url = %config.base_url,
+    };
-                "Using NEAR AI Chat (Responses API, session token auth)"
+    tracing::info!(
-            );
+        model = %config.model,
-            Ok(Arc::new(NearAiProvider::new(config.clone(), session)?))
+        base_url = %config.base_url,
-        }
+        auth = auth_mode,
-        NearAiApiMode::ChatCompletions => {
+        "Using NEAR AI (Chat Completions API)"
-            tracing::info!(
+    );
-                model = %config.model,
+    Ok(Arc::new(NearAiChatProvider::new(config.clone(), session)?))
-                base_url = %config.base_url,
-                "Using NEAR AI Cloud (Chat Completions API, API key auth)"
-            );
-            Ok(Arc::new(NearAiChatProvider::new(config.clone())?))
-        }
-    }
 }
 
 fn create_openai_provider(config: &LlmConfig) -> Result<Arc<dyn LlmProvider>, LlmError> {
@@ -254,7 +246,7 @@ fn create_openai_compatible_provider(config: &LlmConfig) -> Result<Arc<dyn LlmPr
 /// Create a cheap/fast LLM provider for lightweight tasks (heartbeat, routing, evaluation).
 ///
 /// Uses `NEARAI_CHEAP_MODEL` if set, otherwise falls back to the main provider.
-/// Currently only supports NEAR AI backends (Responses and ChatCompletions modes).
+/// Currently only supports NEAR AI backend.
 pub fn create_cheap_llm_provider(
     config: &LlmConfig,
     session: Arc<SessionManager>,
@@ -275,20 +267,16 @@ pub fn create_cheap_llm_provider(
     let mut cheap_config = config.nearai.clone();
     cheap_config.model = cheap_model.clone();
 
-    tracing::info!("Cheap LLM provider: {}", cheap_model);
+    Ok(Some(Arc::new(NearAiChatProvider::new(
-
+        cheap_config,
-    match cheap_config.api_mode {
+        session,
-        NearAiApiMode::Responses => Ok(Some(Arc::new(NearAiProvider::new(cheap_config, session)?))),
+    )?)))
-        NearAiApiMode::ChatCompletions => {
-            Ok(Some(Arc::new(NearAiChatProvider::new(cheap_config)?)))
-        }
-    }
 }
 
 #[cfg(test)]
 mod tests {
     use super::*;
-    use crate::config::{LlmBackend, NearAiApiMode, NearAiConfig};
+    use crate::config::{LlmBackend, NearAiConfig};
     use std::path::PathBuf;
 
     fn test_nearai_config() -> NearAiConfig {
@@ -298,7 +286,6 @@ mod tests {
             base_url: "https://api.near.ai".to_string(),
             auth_base_url: "https://private.near.ai".to_string(),
             session_path: PathBuf::from("/tmp/test-session.json"),
-            api_mode: NearAiApiMode::Responses,
             api_key: None,
             fallback_model: None,
             max_retries: 3,

src/llm/nearai_chat.rs

@@ -1,8 +1,12 @@
-//! NEAR AI Cloud provider implementation (Chat Completions API).
+//! NEAR AI provider implementation (Chat Completions API).
 //!
-//! This provider uses the NEAR AI Cloud API (`cloud-api.near.ai`) which
+//! This provider uses the OpenAI-compatible Chat Completions endpoint with
-//! exposes an OpenAI-compatible chat completions endpoint with API key
+//! dual auth support:
-//! authentication.
+//! - **API key auth**: When `NEARAI_API_KEY` is set, uses Bearer API key
+//! - **Session token auth**: Otherwise, uses `SessionManager` for Bearer session token
+//!   with automatic renewal on 401 errors
+
+use std::sync::Arc;
 
 use async_trait::async_trait;
 use reqwest::Client;
@@ -14,38 +18,51 @@ use serde::{Deserialize, Serialize};
 use crate::config::NearAiConfig;
 use crate::error::LlmError;
 use crate::llm::provider::{
-    ChatMessage, CompletionRequest, CompletionResponse, FinishReason, LlmProvider, ModelMetadata,
+    ChatMessage, CompletionRequest, CompletionResponse, FinishReason, LlmProvider, Role, ToolCall,
-    Role, ToolCall, ToolCompletionRequest, ToolCompletionResponse,
+    ToolCompletionRequest, ToolCompletionResponse,
 };
+use crate::llm::session::SessionManager;
 
-/// NEAR AI Cloud provider (Chat Completions API, API key auth).
+/// Information about an available model from NEAR AI API.
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct ModelInfo {
+    /// Model identifier.
+    #[serde(alias = "id", alias = "model")]
+    pub name: String,
+    /// Optional provider name.
+    #[serde(default)]
+    pub provider: Option<String>,
+}
+
+/// NEAR AI provider (Chat Completions API, dual auth).
 pub struct NearAiChatProvider {
     client: Client,
     config: NearAiConfig,
+    /// Session manager for session token auth (used when no API key is set).
+    session: Arc<SessionManager>,
     active_model: std::sync::RwLock<String>,
     flatten_tool_messages: bool,
 }
 
 impl NearAiChatProvider {
-    /// Create a new NEAR AI Cloud provider with API key auth.
+    /// Create a new NEAR AI Chat Completions provider.
+    ///
+    /// Auth mode is determined by `config.api_key`:
+    /// - If set, uses Bearer API key auth
+    /// - If not set, uses session token auth via `SessionManager`
     ///
     /// By default this enables tool-message flattening for compatibility with
     /// providers that reject `role: "tool"` messages.
-    pub fn new(config: NearAiConfig) -> Result<Self, LlmError> {
+    pub fn new(config: NearAiConfig, session: Arc<SessionManager>) -> Result<Self, LlmError> {
-        Self::new_with_flatten(config, true)
+        Self::new_with_flatten(config, session, true)
     }
 
     /// Create a chat completions provider with configurable tool-message flattening.
     pub fn new_with_flatten(
         config: NearAiConfig,
+        session: Arc<SessionManager>,
         flatten_tool_messages: bool,
     ) -> Result<Self, LlmError> {
-        if config.api_key.is_none() {
-            return Err(LlmError::AuthFailed {
-                provider: "nearai_chat".to_string(),
-            });
-        }
-
         let client = Client::builder()
             .timeout(std::time::Duration::from_secs(120))
             .build()
@@ -58,6 +75,7 @@ impl NearAiChatProvider {
         Ok(Self {
             client,
             config,
+            session,
             active_model,
             flatten_tool_messages,
         })
@@ -74,23 +92,50 @@ impl NearAiChatProvider {
         }
     }
 
-    fn api_key(&self) -> String {
+    /// Returns true if using API key auth, false if session token auth.
-        self.config
+    fn uses_api_key(&self) -> bool {
-            .api_key
+        self.config.api_key.is_some()
-            .as_ref()
+    }
-            .map(|k| k.expose_secret().to_string())
+
-            .unwrap_or_default()
+    /// Resolve the Bearer token for the current auth mode.
+    async fn resolve_bearer_token(&self) -> Result<String, LlmError> {
+        if let Some(ref api_key) = self.config.api_key {
+            Ok(api_key.expose_secret().to_string())
+        } else {
+            let token = self.session.get_token().await?;
+            Ok(token.expose_secret().to_string())
+        }
     }
 
     /// Send a single request to the chat completions API.
     ///
-    /// Does not retry internally — retries are handled by the external
+    /// For session token auth, handles 401 by calling `session.handle_auth_failure()`
+    /// and retrying once.
+    ///
+    /// Does not retry on other errors — retries are handled by the external
     /// `RetryProvider` wrapper in the composition chain.
     async fn send_request<T: Serialize, R: for<'de> Deserialize<'de>>(
         &self,
         body: &T,
+    ) -> Result<R, LlmError> {
+        match self.send_request_inner(body).await {
+            Ok(result) => Ok(result),
+            Err(LlmError::SessionExpired { .. }) if !self.uses_api_key() => {
+                // Session expired, attempt renewal and retry once
+                self.session.handle_auth_failure().await?;
+                self.send_request_inner(body).await
+            }
+            Err(e) => Err(e),
+        }
+    }
+
+    /// Inner request implementation (single attempt).
+    async fn send_request_inner<T: Serialize, R: for<'de> Deserialize<'de>>(
+        &self,
+        body: &T,
     ) -> Result<R, LlmError> {
         let url = self.api_url("chat/completions");
+        let token = self.resolve_bearer_token().await?;
 
         tracing::debug!("Sending request to NEAR AI Chat: {}", url);
 
@@ -103,7 +148,7 @@ impl NearAiChatProvider {
         let response = self
             .client
             .post(&url)
-            .header("Authorization", format!("Bearer {}", self.api_key()))
+            .header("Authorization", format!("Bearer {}", token))
             .header("Content-Type", "application/json")
             .json(body)
             .send()
@@ -126,6 +171,17 @@ impl NearAiChatProvider {
             let status_code = status.as_u16();
 
             if status_code == 401 {
+                // For session token auth, distinguish session expired from plain auth failure
+                if !self.uses_api_key() {
+                    let lower = response_text.to_lowercase();
+                    let is_session_expired = lower.contains("session")
+                        && (lower.contains("expired") || lower.contains("invalid"));
+                    if is_session_expired {
+                        return Err(LlmError::SessionExpired {
+                            provider: "nearai_chat".to_string(),
+                        });
+                    }
+                }
                 return Err(LlmError::AuthFailed {
                     provider: "nearai_chat".to_string(),
                 });
@@ -154,14 +210,31 @@ impl NearAiChatProvider {
         })
     }
 
-    /// Fetch available models with full metadata from the `/v1/models` endpoint.
+    /// Fetch available models from the NEAR AI API.
-    async fn fetch_models(&self) -> Result<Vec<ApiModelEntry>, LlmError> {
+    ///
+    /// Handles session renewal on 401 (same pattern as `send_request`).
+    /// Supports multiple response formats: `{models: [...]}`, `{data: [...]}`, and plain array.
+    pub async fn list_models_full(&self) -> Result<Vec<ModelInfo>, LlmError> {
+        match self.list_models_inner().await {
+            Ok(models) => Ok(models),
+            Err(LlmError::SessionExpired { .. }) if !self.uses_api_key() => {
+                self.session.handle_auth_failure().await?;
+                self.list_models_inner().await
+            }
+            Err(e) => Err(e),
+        }
+    }
+
+    async fn list_models_inner(&self) -> Result<Vec<ModelInfo>, LlmError> {
         let url = self.api_url("models");
+        let token = self.resolve_bearer_token().await?;
+
+        tracing::debug!("Fetching models from: {}", url);
 
         let response = self
             .client
             .get(&url)
-            .header("Authorization", format!("Bearer {}", self.api_key()))
+            .header("Authorization", format!("Bearer {}", token))
             .send()
             .await
             .map_err(|e| LlmError::RequestFailed {
@@ -176,6 +249,11 @@ impl NearAiChatProvider {
         })?;
 
         if !status.is_success() {
+            if status.as_u16() == 401 && !self.uses_api_key() {
+                return Err(LlmError::SessionExpired {
+                    provider: "nearai_chat".to_string(),
+                });
+            }
             let truncated = crate::agent::truncate_for_preview(&response_text, 512);
             return Err(LlmError::RequestFailed {
                 provider: "nearai_chat".to_string(),
@@ -183,29 +261,97 @@ impl NearAiChatProvider {
             });
         }
 
+        // Flexible model entry parsing -- handle various field names
         #[derive(Deserialize)]
-        struct ModelsResponse {
+        struct ModelMetadataInner {
-            data: Vec<ApiModelEntry>,
+            #[serde(default)]
+            name: Option<String>,
+            #[serde(default, alias = "modelName", alias = "model_name")]
+            model_name: Option<String>,
         }
 
-        let resp: ModelsResponse =
+        #[derive(Deserialize)]
-            serde_json::from_str(&response_text).map_err(|e| LlmError::InvalidResponse {
+        struct ModelEntry {
-                provider: "nearai_chat".to_string(),
+            #[serde(default)]
-                reason: format!("JSON parse error: {}", e),
+            name: Option<String>,
-            })?;
+            #[serde(default)]
+            id: Option<String>,
+            #[serde(default)]
+            model: Option<String>,
+            #[serde(default, alias = "modelName", alias = "model_name")]
+            model_name: Option<String>,
+            #[serde(default, alias = "modelId", alias = "model_id")]
+            model_id: Option<String>,
+            #[serde(default)]
+            metadata: Option<ModelMetadataInner>,
+        }
 
-        Ok(resp.data)
+        impl ModelEntry {
+            fn get_name(&self) -> Option<String> {
+                self.name
+                    .clone()
+                    .or_else(|| self.id.clone())
+                    .or_else(|| self.model.clone())
+                    .or_else(|| self.model_name.clone())
+                    .or_else(|| self.model_id.clone())
+                    .or_else(|| self.metadata.as_ref().and_then(|m| m.name.clone()))
+                    .or_else(|| self.metadata.as_ref().and_then(|m| m.model_name.clone()))
+            }
+        }
+
+        #[derive(Deserialize)]
+        struct ModelsResponse {
+            #[serde(default)]
+            models: Option<Vec<ModelEntry>>,
+            #[serde(default)]
+            data: Option<Vec<ModelEntry>>,
+        }
+
+        // Try {models: [...]} or {data: [...]} format
+        if let Ok(resp) = serde_json::from_str::<ModelsResponse>(&response_text)
+            && let Some(entries) = resp.models.or(resp.data)
+        {
+            let models: Vec<ModelInfo> = entries
+                .into_iter()
+                .filter_map(|e| {
+                    e.get_name().map(|name| ModelInfo {
+                        name,
+                        provider: None,
+                    })
+                })
+                .collect();
+            if !models.is_empty() {
+                return Ok(models);
+            }
+        }
+
+        // Try direct array format
+        if let Ok(entries) = serde_json::from_str::<Vec<ModelEntry>>(&response_text) {
+            let models: Vec<ModelInfo> = entries
+                .into_iter()
+                .filter_map(|e| {
+                    e.get_name().map(|name| ModelInfo {
+                        name,
+                        provider: None,
+                    })
+                })
+                .collect();
+            if !models.is_empty() {
+                return Ok(models);
+            }
+        }
+
+        // Couldn't find model names in response
+        Err(LlmError::InvalidResponse {
+            provider: "nearai_chat".to_string(),
+            reason: format!(
+                "No model names found in response: {}",
+                &response_text[..response_text.len().min(300)]
+            ),
+        })
     }
 }
 
-/// Model entry as returned by the `/v1/models` API.
-#[derive(Debug, Deserialize)]
-struct ApiModelEntry {
-    id: String,
-    #[serde(default)]
-    context_length: Option<u32>,
-}
-
 #[async_trait]
 impl LlmProvider for NearAiChatProvider {
     async fn complete(&self, req: CompletionRequest) -> Result<CompletionResponse, LlmError> {
@@ -252,7 +398,6 @@ impl LlmProvider for NearAiChatProvider {
             finish_reason,
             input_tokens,
             output_tokens,
-            response_id: None,
         })
     }
 
@@ -347,7 +492,6 @@ impl LlmProvider for NearAiChatProvider {
             finish_reason,
             input_tokens,
             output_tokens,
-            response_id: None,
         })
     }
 
@@ -361,18 +505,8 @@ impl LlmProvider for NearAiChatProvider {
     }
 
     async fn list_models(&self) -> Result<Vec<String>, LlmError> {
-        let models = self.fetch_models().await?;
+        let models = self.list_models_full().await?;
-        Ok(models.into_iter().map(|m| m.id).collect())
+        Ok(models.into_iter().map(|m| m.name).collect())
-    }
-
-    async fn model_metadata(&self) -> Result<ModelMetadata, LlmError> {
-        let active = self.active_model_name();
-        let models = self.fetch_models().await?;
-        let current = models.iter().find(|m| m.id == active);
-        Ok(ModelMetadata {
-            id: active,
-            context_length: current.and_then(|m| m.context_length),
-        })
     }
 
     fn active_model_name(&self) -> String {
@@ -613,6 +747,7 @@ fn parse_usage(usage: Option<&ChatCompletionUsage>) -> (u32, u32) {
 #[cfg(test)]
 mod tests {
     use super::*;
+    use crate::llm::session::SessionConfig;
 
     fn test_nearai_config(base_url: &str) -> NearAiConfig {
         NearAiConfig {
@@ -620,7 +755,6 @@ mod tests {
             base_url: base_url.to_string(),
             auth_base_url: "https://private.near.ai".to_string(),
             session_path: std::path::PathBuf::from("/tmp/session.json"),
-            api_mode: crate::config::NearAiApiMode::ChatCompletions,
             api_key: Some(secrecy::SecretString::from("test-key".to_string())),
             cheap_model: None,
             fallback_model: None,
@@ -635,18 +769,22 @@ mod tests {
         }
     }
 
+    fn test_session() -> Arc<SessionManager> {
+        Arc::new(SessionManager::new(SessionConfig::default()))
+    }
+
     #[test]
     fn test_api_url_with_base_without_v1() {
         let mut cfg = test_nearai_config("http://127.0.0.1:8318");
 
-        let provider = NearAiChatProvider::new(cfg.clone()).expect("provider");
+        let provider = NearAiChatProvider::new(cfg.clone(), test_session()).expect("provider");
         assert_eq!(
             provider.api_url("chat/completions"),
             "http://127.0.0.1:8318/v1/chat/completions"
         );
 
         cfg.base_url = "http://127.0.0.1:8318/".to_string();
-        let provider = NearAiChatProvider::new(cfg).expect("provider");
+        let provider = NearAiChatProvider::new(cfg, test_session()).expect("provider");
         assert_eq!(
             provider.api_url("/chat/completions"),
             "http://127.0.0.1:8318/v1/chat/completions"
@@ -657,7 +795,7 @@ mod tests {
     fn test_api_url_with_base_already_v1() {
         let cfg = test_nearai_config("http://127.0.0.1:8318/v1");
 
-        let provider = NearAiChatProvider::new(cfg).expect("provider");
+        let provider = NearAiChatProvider::new(cfg, test_session()).expect("provider");
         assert_eq!(
             provider.api_url("chat/completions"),
             "http://127.0.0.1:8318/v1/chat/completions"

src/llm/provider.rs

@@ -153,8 +153,6 @@ pub struct CompletionResponse {
     pub input_tokens: u32,
     pub output_tokens: u32,
     pub finish_reason: FinishReason,
-    /// Provider-specific response ID (e.g. for NEAR AI response chaining).
-    pub response_id: Option<String>,
 }
 
 /// Why the completion finished.
@@ -256,8 +254,6 @@ pub struct ToolCompletionResponse {
     pub input_tokens: u32,
     pub output_tokens: u32,
     pub finish_reason: FinishReason,
-    /// Provider-specific response ID (e.g. for NEAR AI response chaining).
-    pub response_id: Option<String>,
 }
 
 /// Metadata about a model returned by the provider's API.
@@ -327,20 +323,6 @@ pub trait LlmProvider: Send + Sync {
         })
     }
 
-    /// Seed a response chain for a thread (e.g. restoring from DB).
-    ///
-    /// Providers that support response chaining (e.g. NEAR AI `previous_response_id`)
-    /// store this so subsequent calls send only delta messages.
-    fn seed_response_chain(&self, _thread_id: &str, _response_id: String) {}
-
-    /// Get the last response chain ID for a thread.
-    ///
-    /// Returns `None` if the provider doesn't support chaining or has no
-    /// stored state for this thread.
-    fn get_response_chain_id(&self, _thread_id: &str) -> Option<String> {
-        None
-    }
-
     /// Calculate cost for a completion.
     fn calculate_cost(&self, input_tokens: u32, output_tokens: u32) -> Decimal {
         let (input_cost, output_cost) = self.cost_per_token();

src/llm/response_cache.rs

@@ -228,14 +228,6 @@ impl LlmProvider for CachedProvider {
     fn set_model(&self, model: &str) -> Result<(), LlmError> {
         self.inner.set_model(model)
     }
-
-    fn seed_response_chain(&self, thread_id: &str, response_id: String) {
-        self.inner.seed_response_chain(thread_id, response_id);
-    }
-
-    fn get_response_chain_id(&self, thread_id: &str) -> Option<String> {
-        self.inner.get_response_chain_id(thread_id)
-    }
 }
 
 #[cfg(test)]

src/llm/retry.rs

@@ -210,14 +210,6 @@ impl LlmProvider for RetryProvider {
         self.inner.set_model(model)
     }
 
-    fn seed_response_chain(&self, thread_id: &str, response_id: String) {
-        self.inner.seed_response_chain(thread_id, response_id)
-    }
-
-    fn get_response_chain_id(&self, thread_id: &str) -> Option<String> {
-        self.inner.get_response_chain_id(thread_id)
-    }
-
     fn calculate_cost(&self, input_tokens: u32, output_tokens: u32) -> Decimal {
         self.inner.calculate_cost(input_tokens, output_tokens)
     }

src/llm/rig_adapter.rs

@@ -445,7 +445,6 @@ where
             input_tokens: saturate_u32(response.usage.input_tokens),
             output_tokens: saturate_u32(response.usage.output_tokens),
             finish_reason: finish,
-            response_id: None,
         })
     }
 
@@ -511,7 +510,6 @@ where
             input_tokens: saturate_u32(response.usage.input_tokens),
             output_tokens: saturate_u32(response.usage.output_tokens),
             finish_reason: finish,
-            response_id: None,
         })
     }
 

src/llm/session.rs

@@ -513,20 +513,30 @@ impl SessionManager {
         })? {
             value
         } else {
-            tracing::warn!(
+            // Try the legacy key. Only warn if it actually exists (real
-                "nearai.session_token missing; falling back to legacy nearai.session for backwards compatibility"
+            // backwards-compat migration). When neither key is present
-            );
+            // (fresh install), just return the "No session in DB" error.
-            store
+            let legacy = store
                 .get_setting(&user_id, "nearai.session")
                 .await
                 .map_err(|e| LlmError::SessionRenewalFailed {
                     provider: "nearai".to_string(),
                     reason: format!("DB query failed: {}", e),
-                })?
+                })?;
-                .ok_or(LlmError::SessionRenewalFailed {
+            match legacy {
-                    provider: "nearai".to_string(),
+                Some(value) => {
-                    reason: "No session in DB".to_string(),
+                    tracing::warn!(
-                })?
+                        "nearai.session_token missing; falling back to legacy nearai.session for backwards compatibility"
+                    );
+                    value
+                }
+                None => {
+                    return Err(LlmError::SessionRenewalFailed {
+                        provider: "nearai".to_string(),
+                        reason: "No session in DB".to_string(),
+                    });
+                }
+            }
         };
 
         let session: SessionData =

src/main.rs

@@ -3,7 +3,7 @@
 use std::sync::Arc;
 
 use clap::Parser;
-use tracing_subscriber::{EnvFilter, layer::SubscriberExt, util::SubscriberInitExt};
+use tracing_subscriber::EnvFilter;
 
 use ironclaw::{
     agent::{Agent, AgentDeps, SessionManager},
@@ -14,7 +14,7 @@ use ironclaw::{
             RegisteredEndpoint, SharedWasmChannel, WasmChannelLoader, WasmChannelRouter,
             WasmChannelRuntime, WasmChannelRuntimeConfig, create_wasm_channel_router,
         },
-        web::log_layer::{LogBroadcaster, WebLogLayer},
+        web::log_layer::LogBroadcaster,
     },
     cli::{
         Cli, Command, run_mcp_command, run_pairing_command, run_service_command,
@@ -201,6 +201,9 @@ async fn main() -> anyhow::Result<()> {
                 )
                 .init();
 
+            let _ = dotenvy::dotenv();
+            ironclaw::bootstrap::load_ironclaw_env();
+
             return ironclaw::cli::run_doctor_command().await;
         }
         Some(Command::Status) => {
@@ -210,6 +213,9 @@ async fn main() -> anyhow::Result<()> {
                 )
                 .init();
 
+            let _ = dotenvy::dotenv();
+            ironclaw::bootstrap::load_ironclaw_env();
+
             return run_status_command().await;
         }
         Some(Command::Worker {
@@ -360,23 +366,14 @@ async fn main() -> anyhow::Result<()> {
     };
     let session = create_session_manager(session_config).await;
 
-    // Initialize tracing
-    let env_filter = EnvFilter::try_from_default_env()
-        .unwrap_or_else(|_| EnvFilter::new("ironclaw=info,tower_http=warn"));
-
     // Create log broadcaster before tracing init so the WebLogLayer can capture all events.
     // This gets wired to the gateway's /api/logs/events SSE endpoint later.
     let log_broadcaster = Arc::new(LogBroadcaster::new());
 
-    tracing_subscriber::registry()
+    // Initialize tracing with a reloadable EnvFilter so the gateway can switch
-        .with(env_filter)
+    // log levels (e.g. ironclaw=debug) at runtime without restarting.
-        .with(
+    let log_level_handle =
-            tracing_subscriber::fmt::layer()
+        ironclaw::channels::web::log_layer::init_tracing(Arc::clone(&log_broadcaster));
-                .with_target(false)
-                .with_writer(ironclaw::tracing_fmt::TruncatingStderr::default()),
-        )
-        .with(WebLogLayer::new(Arc::clone(&log_broadcaster)))
-        .init();
 
     // Create CLI channel
     let repl_channel = if let Some(ref msg) = cli.message {
@@ -730,7 +727,6 @@ async fn main() -> anyhow::Result<()> {
     // Initialize tool registry
     let tools = Arc::new(ToolRegistry::new());
     tools.register_builtin_tools();
-    tracing::info!("Registered {} built-in tools", tools.count());
 
     // Create embeddings provider if configured
     let embeddings: Option<Arc<dyn EmbeddingProvider>> = if config.embeddings.enabled {
@@ -1024,11 +1020,12 @@ async fn main() -> anyhow::Result<()> {
     // Set up orchestrator for sandboxed job execution
     // When allow_local_tools is false (default), the LLM uses create_job for FS/shell work.
     // When allow_local_tools is true, dev tools are also registered directly (current behavior).
-    if config.agent.allow_local_tools {
+    // register_builder_tool() already calls register_dev_tools() internally,
+    // so only register them here when the builder didn't already do it.
+    let builder_registered_dev_tools =
+        config.builder.enabled && (config.agent.allow_local_tools || !config.sandbox.enabled);
+    if config.agent.allow_local_tools && !builder_registered_dev_tools {
         tools.register_dev_tools();
-        tracing::info!(
-            "Local tools enabled (allow_local_tools=true), dev tools registered directly"
-        );
     }
 
     // Shared state for job events (used by both orchestrator and web gateway)
@@ -1079,7 +1076,6 @@ async fn main() -> anyhow::Result<()> {
             }
         });
 
-        tracing::info!("Orchestrator API started on :50051, sandbox delegation enabled");
         if config.claude_code.enabled {
             tracing::info!(
                 "Claude Code sandbox mode available (model: {}, max_turns: {})",
@@ -1333,9 +1329,6 @@ async fn main() -> anyhow::Result<()> {
     // Seed workspace with core identity files on first boot
     if let Some(ref ws) = workspace {
         match ws.seed_if_empty().await {
-            Ok(count) if count > 0 => {
-                tracing::info!("Workspace seeded with {} core files", count);
-            }
             Ok(_) => {}
             Err(e) => {
                 tracing::warn!("Failed to seed workspace: {}", e);
@@ -1426,6 +1419,7 @@ async fn main() -> anyhow::Result<()> {
         }
         gw = gw.with_session_manager(Arc::clone(&session_manager));
         gw = gw.with_log_broadcaster(Arc::clone(&log_broadcaster));
+        gw = gw.with_log_level_handle(Arc::clone(&log_level_handle));
         gw = gw.with_tool_registry(Arc::clone(&tools));
         if let Some(ref ext_mgr) = extension_manager {
             gw = gw.with_extension_manager(Arc::clone(ext_mgr));
@@ -1464,11 +1458,6 @@ async fn main() -> anyhow::Result<()> {
             gw.auth_token()
         ));
 
-        tracing::info!(
-            "Web gateway enabled on {}:{}",
-            gw_config.host,
-            gw_config.port
-        );
         tracing::info!("Web UI: http://{}:{}/", gw_config.host, gw_config.port);
 
         channel_names.push("gateway".to_string());
@@ -1511,8 +1500,6 @@ async fn main() -> anyhow::Result<()> {
         Some(session_manager),
     );
 
-    tracing::info!("Agent initialized, starting main loop...");
-
     // Print boot screen for interactive CLI mode (not single-message mode).
     if config.channels.cli.enabled && cli.message.is_none() {
         let boot_info = ironclaw::boot_screen::BootInfo {

src/setup/wizard.rs

@@ -826,6 +826,12 @@ impl SetupWizard {
 
         self.session_manager = Some(session);
 
+        // Persist session token to the database so the runtime can load it
+        // via `attach_store()` → `load_session_from_db()` without the
+        // backwards-compat fallback. The session manager saved to disk but
+        // doesn't have a DB store attached during onboarding.
+        self.persist_session_to_db().await;
+
         // If the user chose the API key path, NEARAI_API_KEY is now set
         // in the environment. Persist it to the encrypted secrets store
         // so inject_llm_keys_from_secrets() can load it on future runs.
@@ -1160,7 +1166,6 @@ impl SetupWizard {
                 base_url,
                 auth_base_url,
                 session_path: crate::llm::session::default_session_path(),
-                api_mode: crate::config::NearAiApiMode::Responses,
                 api_key: None,
                 fallback_model: None,
                 max_retries: 3,
@@ -1861,6 +1866,54 @@ impl SetupWizard {
         Ok(())
     }
 
+    /// Persist the NEAR AI session token to the database.
+    ///
+    /// The session manager writes to disk during `ensure_authenticated()` but
+    /// doesn't have a DB store attached during onboarding. This reads the
+    /// session file from disk and stores it under the `nearai.session_token`
+    /// key so the runtime's `attach_store()` finds it without fallback.
+    ///
+    /// Best-effort: silently ignores errors (no DB connection yet, no
+    /// session file, etc.).
+    async fn persist_session_to_db(&self) {
+        let session_path = crate::llm::session::default_session_path();
+        let data = match std::fs::read_to_string(&session_path) {
+            Ok(d) if !d.trim().is_empty() => d,
+            _ => return,
+        };
+        let value: serde_json::Value = match serde_json::from_str(&data) {
+            Ok(v) => v,
+            Err(_) => return,
+        };
+
+        #[cfg(feature = "postgres")]
+        if let Some(ref pool) = self.db_pool {
+            let store = crate::history::Store::from_pool(pool.clone());
+            if let Err(e) = store
+                .set_setting("default", "nearai.session_token", &value)
+                .await
+            {
+                tracing::debug!("Could not persist session token to postgres: {}", e);
+            } else {
+                tracing::debug!("Session token persisted to database");
+                return;
+            }
+        }
+
+        #[cfg(feature = "libsql")]
+        if let Some(ref backend) = self.db_backend {
+            use crate::db::SettingsStore as _;
+            if let Err(e) = backend
+                .set_setting("default", "nearai.session_token", &value)
+                .await
+            {
+                tracing::debug!("Could not persist session token to libsql: {}", e);
+            } else {
+                tracing::debug!("Session token persisted to database");
+            }
+        }
+    }
+
     /// Persist settings to DB and bootstrap .env after each step.
     ///
     /// Silently ignores errors (e.g., DB not connected yet before step 1

src/skills/registry.rs

@@ -197,7 +197,7 @@ impl SkillRegistry {
                     let source = make_source(path.clone());
                     match self.load_skill_md(&skill_md, trust, source).await {
                         Ok((name, skill)) => {
-                            tracing::info!("Loaded skill: {}", name);
+                            tracing::debug!("Loaded skill: {}", name);
                             results.push((name, skill));
                         }
                         Err(e) => {

src/testing.rs

@@ -168,7 +168,6 @@ impl LlmProvider for StubLlm {
             input_tokens: 10,
             output_tokens: 5,
             finish_reason: FinishReason::Stop,
-            response_id: None,
         })
     }
 
@@ -186,7 +185,6 @@ impl LlmProvider for StubLlm {
             input_tokens: 10,
             output_tokens: 5,
             finish_reason: FinishReason::Stop,
-            response_id: None,
         })
     }
 }

src/worker/api.rs

@@ -227,7 +227,6 @@ impl WorkerHttpClient {
             input_tokens: proxy_resp.input_tokens,
             output_tokens: proxy_resp.output_tokens,
             finish_reason: parse_finish_reason(&proxy_resp.finish_reason),
-            response_id: None,
         })
     }
 
@@ -255,7 +254,6 @@ impl WorkerHttpClient {
             input_tokens: proxy_resp.input_tokens,
             output_tokens: proxy_resp.output_tokens,
             finish_reason: parse_finish_reason(&proxy_resp.finish_reason),
-            response_id: None,
         })
     }
 

tests/heartbeat_integration.rs

@@ -15,6 +15,7 @@ use ironclaw::{
     config::Config,
     history::Store,
     llm::{SessionConfig, create_llm_provider, create_session_manager},
+    safety::SafetyLayer,
     workspace::Workspace,
 };
 
@@ -96,7 +97,8 @@ async fn test_heartbeat_end_to_end() {
 
     let hb_config = ironclaw::agent::HeartbeatConfig::default();
     let hygiene_config = ironclaw::workspace::hygiene::HygieneConfig::default();
-    let runner = HeartbeatRunner::new(hb_config, hygiene_config, workspace, llm);
+    let safety = Arc::new(SafetyLayer::new(&config.safety));
+    let runner = HeartbeatRunner::new(hb_config, hygiene_config, workspace, llm, safety);
 
     let result = runner.check_heartbeat().await;
 

tests/openai_compat_integration.rs

@@ -71,7 +71,6 @@ impl LlmProvider for MockLlmProvider {
             input_tokens: 10,
             output_tokens: 5,
             finish_reason: FinishReason::Stop,
-            response_id: None,
         })
     }
 
@@ -97,7 +96,6 @@ impl LlmProvider for MockLlmProvider {
                 input_tokens: 15,
                 output_tokens: 8,
                 finish_reason: FinishReason::ToolUse,
-                response_id: None,
             })
         } else {
             Ok(ToolCompletionResponse {
@@ -106,7 +104,6 @@ impl LlmProvider for MockLlmProvider {
                 input_tokens: 10,
                 output_tokens: 4,
                 finish_reason: FinishReason::Stop,
-                response_id: None,
             })
         }
     }
@@ -145,7 +142,6 @@ impl LlmProvider for FixedModelProvider {
             input_tokens: 10,
             output_tokens: 5,
             finish_reason: FinishReason::Stop,
-            response_id: None,
         })
     }
 
@@ -159,7 +155,6 @@ impl LlmProvider for FixedModelProvider {
             input_tokens: 10,
             output_tokens: 5,
             finish_reason: FinishReason::Stop,
-            response_id: None,
         })
     }
 
@@ -190,6 +185,7 @@ async fn start_test_server_with_provider(
         workspace: None,
         session_manager: None,
         log_broadcaster: None,
+        log_level_handle: None,
         extension_manager: None,
         tool_registry: None,
         store: None,
@@ -674,6 +670,7 @@ async fn test_no_llm_provider_returns_503() {
         workspace: None,
         session_manager: None,
         log_broadcaster: None,
+        log_level_handle: None,
         extension_manager: None,
         tool_registry: None,
         store: None,

tests/ws_gateway_integration.rs

@@ -43,6 +43,7 @@ async fn start_test_server() -> (
         workspace: None,
         session_manager: None,
         log_broadcaster: None,
+        log_level_handle: None,
         extension_manager: None,
         tool_registry: None,
         store: None,