菜单
How to handle rate-limit emergencies and the FATAL cascade when the bridge goes blind. Includes the cooldown-bypass triggers, the L4-SOFT/L5-HARD recovery path, and the RESET SESSIONE handling on a >30-point usage drop.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Extract the company logo for a company in the companies table and store it as a small base64 data-URI (max ~35KB, min 32px). Primary path is fully automated via logo_fetch.py against the official website (apple-touch-icon → icon → og:image → favicon); when the site blocks bots or has no usable icon, find a direct logo image URL via web search and pass it with --from-url. Verify the website really belongs to the company BEFORE fetching. Sets companies.logo, logo_source, logo_fetched.
Lo sweep di manutenzione INFRA del Mantenitore 👷♂️ (gemello del Dottore, scope infrastruttura non agenti). Una passata giornaliera one-shot: canary di liveness dei processi salva-vita del container (bridge/daemon/watchdog) via process_health.py, smoke-test dei tool mission-critical (browser/LinkedIn) via tool_health.py, audit/consolidamento deps fuori standard, GC di script e tmp orfani, de-dup di script ricorrenti, freschezza deps, trend disco/RAM. Single-writer: il Mantenitore è l'UNICO che ripara l'infra; le azioni DISTRUTTIVE (delete/archive) le PROPONE, il Capitano decide. Esito in append su mantenitore-logbook.jsonl.
Doctor-only. Context-refresh round: for each agent session read its real context occupancy (provider client-side command, zero tokens) and refresh ONLY sessions whose context window is >50% full — do a retrospective (capture + interview + analytics), append a dense synthesis to the growing daily journal, then KILL + recreate + resume the session with continuation context, so its context window is cleared without losing where it was. Runs 2× per work window (at +30min and at mid). Skips fresh, low-context (≤50%), and Capitano-parked sessions.
Run the mandatory 3-round CV review loop with the Critico — autonomously, without going through the Capitano. For each round you spawn a FRESH `CRITICO-S<N>` session (same N as your Scrittore session: SCRITTORE-2 → CRITICO-S2), send PDF + JD, wait for the structured verdict, kill the Critic, correct the CV, regenerate the PDF, and start the next round with another fresh instance. Three rounds are non-negotiable — neither 1 nor 2. After the 3rd round, gate: `critic_score ≥ 5` → `ready`, else `excluded`. Owned by the Scrittore.
DB + filesystem contract every Scrittore follows when taking a position from `scored` (≥50) to `ready`/`excluded`. Three gates BEFORE writing a single line of CV (anti-rewriting, anti-collision, link verification), one canonical path for deliverables, one final gate after the 3rd Critic round. Skipping any of these produces duplicate work, overwrites another Writer's claim, or — worst — pushes an `excluded`-grade CV to the user as `ready`. Owned by the Scrittore.
Insert NEW records into the JHT DB (positions / scores / applications / companies / position_highlights). Use it ONLY when an agent needs to create a record — Scout for positions, Analyst for companies and highlights, Scorer for scores, Writer for applications. Never blind-overwrite — for updates use `db-update`.
| name | emergency-handling |
| description | How to handle rate-limit emergencies and the FATAL cascade when the bridge goes blind. Includes the cooldown-bypass triggers, the L4-SOFT/L5-HARD recovery path, and the RESET SESSIONE handling on a >30-point usage drop. |
| allowed-tools | Bash(python3 *) |
Una di queste condizioni → manda ordine immediato senza aspettare cooldown:
proj > 200% (catastrofica)velocità_smussata > velocità_ideale × 5 (esplosione)usage ≥ 90% assoluto (limite hard)In questi casi, PRIMA della notifica esegui freeze_team.py:
python3 /app/shared/skills/freeze_team.py
Manda Esc x2 a tutti gli operativi (esclude CAPITANO/ASSISTENTE/SENTINELLA/SENTINELLA-WORKER). Il consumo si ferma anche se il messaggio al Capitano si perde.
Imposta freeze_active = True.
Mantieni emergency_proj_history (ultimi 5) e emergency_proj_min. Tre trigger:
SE recovery_tracking_cooldown == 0 AND len(history) >= 3:
delta_3 = history[-3] - history[-1]
SE delta_3 > 0: manda RECOVERY TRACKING (calo)
SE delta_3 ≈ 0: → vedi STAGNAZIONE
SE delta_3 < -5: → vedi PEGGIORAMENTO
recovery_tracking_cooldown = 3
SE len(history) >= 5 AND proj > 150% AND (max(history) - min(history)) < 10:
manda STAGNAZIONE CRITICA → "kill altri agenti, throttle non basta"
cooldown 5 tick prima di rimandarla
SE proj > emergency_proj_min + 10:
manda PEGGIORAMENTO POST-FREEZE → "secondo freeze + kill totale"
no cooldown: scatta subito
Quando il bridge non riesce a leggere usage e ricevi [BRIDGE FAILURE]:
L1 — fetch HTTP rapido (vedi skill `check-usage-http`)
• OK → continua normalmente
• FAIL → ↓
L2 — TUI worker manuale (vedi skill `check-usage-tui`)
• OK → continua normalmente
• FAIL → ↓
L3 — FATAL: niente dato dal bridge per N cicli consecutivi
fatal_streak == 0 → 1)python3 /app/shared/skills/soft_pause_team.py
La skill manda 2 messaggi differenziati via jht-tmux-send:
Imposta fatal_streak = 1. Taci finché non arriva BRIDGE TICK valido o INFO.
fatal_streak == 1 → 2)python3 /app/shared/skills/freeze_team.py
Manda Esc x2 a tutti gli operativi (più aggressivo). Inoltre manda al Capitano l'ordine HARD FREEZE (vedi skill order-formats).
Imposta fatal_streak = 2.
Quando arriva un [BRIDGE TICK] valido o [BRIDGE INFO] con fatal_streak >= 1:
fatal_streak = 0, freeze_active = Falseorder-formats)[RIPRENDI] ai suoi operativifatal_streak | Trigger | Azione |
|---|---|---|
| 0 → 1 | primo L1+L2 ko | soft_pause_team.py + PAUSA TEAM al Capitano |
| 1 → 2 | secondo L1+L2 ko consecutivo | freeze_team.py + HARD FREEZE al Capitano |
| ≥ 1 → 0 | [BRIDGE TICK] valido o [BRIDGE INFO] | [RIPRENDI] al Capitano |
Se in un tick rilevi usage sceso di > 30 punti rispetto al sample precedente, è un reset finestra:
memory-state)order-formats)