hive.error-recovery

1## Operational Protocol: Error Recovery
2 
3When a tool call fails:
4 
51. **Diagnose** — classify the failure as *transient* (network blip, rate limit, timeout) or *structural* (wrong selector, missing auth, invalid schema, permission denied).
6 
72. **Decide:**
8   - Transient → retry once.
9   - Structural + fixable → fix the input and retry.
10   - Structural + unfixable → record the failure and move to the next item.
11   - Blocking all progress → escalate.
12 
133. **Adapt** — if the same tool has failed {{max_retries_per_tool}}+ times in a row, stop using it and find an alternative approach.
14 
15**Never silently drop a failed item.** If the item is a task in the colony queue, write the failure to the DB instead of an in-memory buffer:
16 
17```bash
18sqlite3 "$DB_PATH" "UPDATE tasks SET status='failed', last_error='<one-sentence reason>', completed_at=datetime('now'), updated_at=datetime('now') WHERE id='<task-id>' AND worker_id='<your-worker-id>';"
19```
20 
21The `tasks.retry_count` column and the stale-claim reclaimer handle auto-retry for crashes; your job is the within-run decision tree above. See `hive.colony-progress-tracker` for the full queue protocol.