Run #3431

{
  "ok": true,
  "failed": [],
  "hallucinated": [],
  "context_errors": []
}

{
  "pool_status": "see_db"
}

Tu es un agent de monitoring du pool agencecom.

Lessons pertinentes:
LESSONS APPRISES PERTINENTES :
- L453-HUB: Health endpoints K8s : separer liveness no-DB / readiness DB-1s alert / startup cached
- L328-CROSS-S150-blocD2-deploy: Bloc D2 cache layer DEPLOYED Sprint 2 P2 — REUSE Redis 7.0.15 BSD existing Contabo (skip Valkey doublon) + smoke 5/5 PASS via /api/memory/cache/health endpoint dedie
- L-S199-LL2-V52-SPEC-REVIEW: L-S199-LL2 — Review V5.2 Monitoring spec (3 BLOCKERS + 15 medium + 5 low = 23 findings, GO sous condition)
- L-S199-CROSS-V52-SPEC-DRAFTED: L-S199-CROSS-V52-SPEC-DRAFTED — V5.2 Monitoring event-driven spec push DB
- L-S189-LL2-observateur-recap: L-S189-LL2 — Observateur PHASE 1 recap 13+ cycles SILENT + cron clôture

Donnees pool actuelles:
see_db

CONTRAINTE STRICTE -- Tables autorisees UNIQUEMENT :
- agencecom.agents (id, name, agent_level, status, current_session_id, error_count_last_hour, total_sessions, total_errors, total_cost_usd, quarantined_until, acquired_at, last_released_at)
- agencecom.agent_task_runs (id, agent_id, status, started_at, completed_at, error_message, llm_tokens_in, llm_tokens_out, cost_usd, validation_result)

Toute autre table (v_agents_critical, agent_pool, healthcheck_config, error_log, file_locks, zombie_agents, agent_audit, v_agent_timeout_health, v_agents_timeout_risk, agent_health_log, redis_health_ping, auto_quarantine, missions, health_checks, agent_health, internal) NEXISTE PAS -- ne reference jamais de table absente de cette liste.

Identifie les agents zombies (busy >1h sans completion via started_at + status), error_count_last_hour > 5, ou drift quarantine (quarantined_until > now).
Donne 3 recommandations concretes en markdown bullet points avec actions SQL utilisant SEULEMENT les 2 tables autorisees ci-dessus.

## 🔍 Diagnostic pool **agencecom** – Observateur (Phase 1, cycle silent)

### 1. Agents zombies (busy > 1 h sans completion)
Les sessions bloquées en `running` depuis plus d’une heure bloquent la capacité du pool.  
**Action SQL de détection** :
```sql
SELECT a.id, a.name, r.id AS run_id, r.started_at,
       EXTRACT(EPOCH FROM (NOW() - r.started_at))/3600 AS heures_blocage
FROM agencecom.agents a
JOIN agencecom.agent_task_runs r
  ON r.agent_id = a.id AND r.id = a.current_session_id
WHERE a.status = 'busy'
  AND r.status = 'running'
  AND r.started_at < NOW() - INTERVAL '1 hour'
  AND r.completed_at IS NULL;
```
**Remédiation** – forcer l’échec de la run zombie et libérer l’agent :
```sql
UPDATE agencecom.agent_task_runs
   SET status = 'failed',
       error_message = 'Zombie timeout (busy >1h)',
       completed_at = NOW()
WHERE id IN (<zombie_run_ids>);

UPDATE agencecom.agents
   SET status = 'available',
       current_session_id = NULL,
       acquired_at = NULL
WHERE id IN (<zombie_agent_ids>);
```

---

### 2. Agents à fort taux d’erreur (error_count_last_hour > 5)
Un agent qui cumule plus de 5 erreurs en une heure doit être isolé pour protéger le flux.  
**Détection** :
```sql
SELECT id, name, error_count_last_hour, status
FROM agencecom.agents
WHERE error_count_last_hour > 5;
```
**Remédiation** – mise en quarantaine automatique (respecter le cycle cron observateur) :
```sql
UPDATE agencecom.agents
   SET quarantined_until = NOW() + INTERVAL '10 minutes',
       status = 'quarantined'
WHERE id IN (<agent_ids>)
  AND (quarantined_until IS NULL OR quarantined_until < NOW());
```

---

### 3. Dérive de quarantaine (quarantined_until > now)
Des agents restent confinés alors qu’ils n’ont plus d’erreurs récentes ; le flag `quarantined_until` dérive dans le futur.  
**Repérage** :
```sql
SELECT id, name, quarantined_until, error_count_last_hour
FROM agencecom.agents
WHERE quarantined_until > NOW();
```
**Remédiation** – levée de quarantaine pour les agents sains et inactifs :
```sql
UPDATE agencecom.agents
   SET quarantined_until = NOW() - INTERVAL '1 second',
       status = 'available'
WHERE id IN (<agent_ids>)
  AND error_count_last_hour = 0
  AND NOT EXISTS (
    SELECT 1 FROM agencecom.agent_task_runs
    WHERE agent_id = agencecom.agents.id AND status = 'running'
  );
```

> ⚠️ Toutes les actions ci‑dessus n’utilisent **que** `agencecom.agents` et `agencecom.agent_task_runs`, conformément à la contrainte des tables autorisées. Le monitoring reste strictement adossé aux deux tables réelles du pool.