feat: disable subagent summaries, label subagent observations (#2073)

* feat: disable subagent summaries and label subagent observations

Detect Claude Code subagent hook context via `agent_id`/`agent_type` on
stdin, short-circuit the Stop-hook summary path when present, and thread
the subagent identity end-to-end onto observation rows (new `agent_type`
and `agent_id` columns, migration 010 at version 27). Main-session rows
remain NULL; content-hash dedup is unchanged.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix: address PR #2073 review feedback

- Narrow summarize subagent guard to agentId only so --agent-started
  main sessions still own their summary (agentType alone is main-session).
- Remove now-dead agentId/agentType spreads from the summarize POST body.
- Always overwrite pendingAgentId/pendingAgentType in SDK/Gemini/OpenRouter
  agents (clears stale subagent identity on main-session messages after
  a subagent message in the same batch).
- Add idx_observations_agent_id index in migration 010 + the mirror
  migration in SessionStore + the runner.
- Replace console.log in migration010 with logger.debug.
- Update summarize test: agentType alone no longer short-circuits.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* fix: address CodeRabbit + claude-review iteration 4 feedback

- SessionRoutes.handleSummarizeByClaudeId: narrow worker-side guard to
  agentId only (matches hook-side). agentType alone = --agent main
  session, which still owns its summary.
- ResponseProcessor: wrap storeObservations in try/finally so
  pendingAgentId/Type clear even if storage throws. Prevents stale
  subagent identity from leaking into the next batch on error.
- SessionStore.importObservation + bulk.importObservation: persist
  agent_type/agent_id so backup/import round-trips preserve subagent
  attribution.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* polish: claude-review iteration 5 cleanup

- Use ?? not || for nullable subagent fields in PendingMessageStore
  (prevents treating empty string as null).
- Simplify observation.ts body spread — include fields unconditionally;
  JSON.stringify drops undefined anyway.
- Narrow any[] to Array<{ name: string }> in migration010 column checks.
- Add trailing newline to migrations.ts.
- Document in observations/store.ts why the dedup hash intentionally
  excludes agent fields.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* polish: claude-review iteration 7 feedback

- claude-code adapter: add 128-char safety cap on agent_id/agent_type
  so a malformed Claude Code payload cannot balloon DB rows. Empty
  strings now also treated as absent.
- migration010: state-aware debug log lists only columns actually
  added; idempotent re-runs log "already present; ensured indexes".
- Add 3 adapter tests covering the length cap boundary and empty-string
  rejection.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* perf: skip subagent summary before worker bootstrap

Move the agentId short-circuit above ensureWorkerRunning() so a Stop
hook fired inside a subagent does not trigger worker startup just to
return early. Addresses CodeRabbit nit on summarize.ts:36-47.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/cli/adapters/claude-code.ts

@@ -2,6 +2,13 @@ import type { PlatformAdapter, NormalizedHookInput, HookResult } from '../types.
 
 // Maps Claude Code stdin format (session_id, cwd, tool_name, etc.)
 // SessionStart hooks receive no stdin, so we must handle undefined input gracefully
+
+// Defensive cap: Claude Code's agent identifiers are short (e.g., "agent-abc123", "Explore").
+// Ignore anything longer than 128 chars so a malformed payload cannot balloon DB rows.
+const MAX_AGENT_FIELD_LEN = 128;
+const pickAgentField = (v: unknown): string | undefined =>
+  typeof v === 'string' && v.length > 0 && v.length <= MAX_AGENT_FIELD_LEN ? v : undefined;
+
 export const claudeCodeAdapter: PlatformAdapter = {
   normalizeInput(raw) {
     const r = (raw ?? {}) as any;
@@ -13,6 +20,8 @@ export const claudeCodeAdapter: PlatformAdapter = {
       toolInput: r.tool_input,
       toolResponse: r.tool_response,
       transcriptPath: r.transcript_path,
+      agentId: pickAgentField(r.agent_id),
+      agentType: pickAgentField(r.agent_type),
     };
   },
   formatOutput(result) {

src/cli/handlers/observation.ts

@@ -57,7 +57,9 @@ export const observationHandler: EventHandler = {
           tool_name: toolName,
           tool_input: toolInput,
           tool_response: toolResponse,
-          cwd
+          cwd,
+          agentId: input.agentId,
+          agentType: input.agentType
         })
       });
 

src/cli/handlers/summarize.ts

@@ -26,6 +26,20 @@ const MAX_WAIT_FOR_SUMMARY_MS = 110_000; // 110s — fits within Stop hook's 120
 
 export const summarizeHandler: EventHandler = {
   async execute(input: NormalizedHookInput): Promise<HookResult> {
+    // Skip summaries in subagent context — subagents do not own the session summary.
+    // Gate on agentId only: that field is present exclusively for Task-spawned subagents.
+    // agentType alone (no agentId) indicates `--agent`-started main sessions, which still
+    // own their summary. Do this BEFORE ensureWorkerRunning() so a subagent Stop hook
+    // does not bootstrap the worker.
+    if (input.agentId) {
+      logger.debug('HOOK', 'Skipping summary: subagent context detected', {
+        sessionId: input.sessionId,
+        agentId: input.agentId,
+        agentType: input.agentType
+      });
+      return { continue: true, suppressOutput: true, exitCode: HOOK_EXIT_CODES.SUCCESS };
+    }
+
     // Ensure worker is running before any other logic
     const workerReady = await ensureWorkerRunning();
     if (!workerReady) {

src/cli/types.ts

@@ -12,6 +12,10 @@ export interface NormalizedHookInput {
   edits?: unknown[];   // afterFileEdit
   // Platform-specific metadata (source, reason, trigger, mcp_context, etc.)
   metadata?: Record<string, unknown>;
+  // Claude Code subagent identity — present only when hook fires inside a subagent.
+  // Main session has both undefined. Discriminator for subagent context.
+  agentId?: string;      // Claude Code subagent agent_id (undefined in main session)
+  agentType?: string;    // Claude Code subagent agent_type (undefined in main session)
 }
 
 export interface HookResult {