agent-status

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.

/agent-status — Agent Status Dashboard

One screen. Health, queue, throughput, recent failures. No spelunking.

Prerequisites

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 summary
• GET /api/tasks?status=running — running tasks
• GET /api/tasks?status=queued — queued tasks
• GET /api/tasks?status=failed&since=24h — recent failures

If your agent exposes different endpoints, adapt the curl invocations below.

When invoked

Pull a single dashboard frame answering:

1. Health — service up? DB reachable? Workers responsive?
2. Active work — what's running right now (count + brief task summaries)
3. Queue — how many queued, how long is the oldest
4. Throughput — tasks completed in the last hour / 24h
5. Failures — fail rate over last 24h, top 3 most recent failures
6. Cost — if the agent tracks per-task spend, sum last 24h

Workflow

# 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"

Output format

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>
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

When to recommend follow-up

• Fail rate >5% over last 24h → suggest /agent-triage
• Queue depth >20 with oldest >1h → suggest investigation (worker stuck?)
• Health degraded → suggest checking logs / restart
• Workers <expected count → suggest checking process supervisor

Notes

• The skill assumes a single agent. For multi-agent setups, set AGENT_API_URL per-invocation or run the skill once per agent.
• Don't paginate failures here — that's /agent-triage's job. Show only the top 3.
• Keep the dashboard one screen. If anything overflows, summarize and suggest the deeper-dive skill.