菜单
At-a-glance overview of an autonomous coding agent — health, running tasks, queue, fail rate. Use to answer "is the agent working and what's it doing?" in a single dashboard.
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
Purge tasks from an autonomous coding agent that failed due to infrastructure errors (rate limits, credit exhaustion, auth failures, network) — not code failures. Cleans the queue without losing legitimate failures that need triage.
Full-cycle improvement loop for an autonomous coding agent. Verify operational health, analyze pipeline state, research cutting-edge agent patterns, propose and apply improvements. The self-upgrading loop.
Deep dive into a single task in an autonomous coding agent — details, log, events, dependencies, review verdict. Use when a task's behavior is mysterious or `/agent-triage` couldn't classify it.
Operational actions on tasks in an autonomous coding agent — stop, requeue, dead-letter, retry-subtasks, bump-priority. The mechanic's toolbox for running agents.
Triage orphan tasks in an autonomous coding agent — tasks marked "done" with commits or branches that never merged to main. Classifies each, shows the diff, and offers bulk merge / dead-letter / requeue dispositions.
Classify failed tasks from an autonomous coding agent, recommend disposition (retry / dead-letter / fix-and-retry / wontfix), and offer bulk action. Replaces reading hundreds of failure logs by hand.
| name | agent-status |
| description | At-a-glance overview of an autonomous coding agent — health, running tasks, queue, fail rate. Use to answer "is the agent working and what's it doing?" in a single dashboard. |
One screen. Health, queue, throughput, recent failures. No spelunking.
Set these env vars (or pass via shell when invoking):
| Var | Purpose |
|---|---|
AGENT_API_URL | Base URL of the agent's HTTP API (e.g. https://agent.example.com or http://localhost:7777) |
AGENT_API_TOKEN | Bearer token for authenticated endpoints |
The skill expects a task-queue-shaped agent that exposes endpoints roughly matching:
GET /api/health — health summaryGET /api/tasks?status=running — running tasksGET /api/tasks?status=queued — queued tasksGET /api/tasks?status=failed&since=24h — recent failuresIf your agent exposes different endpoints, adapt the curl invocations below.
Pull a single dashboard frame answering:
# 1. Health
curl -s -H "Authorization: Bearer $AGENT_API_TOKEN" "$AGENT_API_URL/api/health"
# 2. Running tasks
curl -s -H "Authorization: Bearer $AGENT_API_TOKEN" "$AGENT_API_URL/api/tasks?status=running"
# 3. Queue depth + oldest
curl -s -H "Authorization: Bearer $AGENT_API_TOKEN" "$AGENT_API_URL/api/tasks?status=queued&order=created_at"
# 4. Recent failures
curl -s -H "Authorization: Bearer $AGENT_API_TOKEN" "$AGENT_API_URL/api/tasks?status=failed&since=24h"
AGENT STATUS — <ISO timestamp>
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
HEALTH [✓ healthy | ⚠ degraded | ✗ down]
service: <up/down>
workers: <N/M alive>
db: <reachable/error>
ACTIVE <N running>
• <task-id> <slug> · running <duration>
• <task-id> <slug> · running <duration>
• <task-id> <slug> · running <duration>
QUEUE <N queued | oldest <duration> ago>
LAST 24H <N completed | <N failed> | <X.X% fail rate>
FAILURES (last 3)
• <task-id> <slug> — <error class> — <duration> ago
• <task-id> <slug> — <error class> — <duration> ago
• <task-id> <slug> — <error class> — <duration> ago
COST $<amount> spent / 24h (if tracked)
NEXT Recommended action: <triage failures / clear queue / nothing>
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
/agent-triageAGENT_API_URL per-invocation or run the skill once per agent./agent-triage's job. Show only the top 3.