القائمة
Monitor and manage your Clawdbot agent fleet from within an agent. Query health, sessions, usage, and trigger actions via the Clawd Control dashboard API.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | clawd-control |
| description | Monitor and manage your Clawdbot agent fleet from within an agent. Query health, sessions, usage, and trigger actions via the Clawd Control dashboard API. |
| version | 1.0.0 |
| author | temaki |
| tags | ["dashboard","monitoring","fleet","management","multi-agent"] |
| requires | ["web_fetch"] |
Monitor and manage your Clawdbot agent fleet through the Clawd Control dashboard API. This skill lets agents check fleet health, inspect sessions, trigger heartbeats, and manage other agents — all programmatically.
http://<host>:3100)auth.json in the Clawd Control directoryAll API calls require a session cookie. Obtain one first:
curl -s -c cookies.txt http://<HOST>:3100/api/login \
-H "Content-Type: application/json" \
-d '{"password": "<PASSWORD>"}'
Or use the Authorization: Bearer <session_token> header.
For agents with web_fetch, you can use the session cookie approach or pass the token as a header.
GET /api/snapshot Returns the full fleet state: all agents + host metrics in a single call.
Response shape:
{
"ts": 1706000000000,
"agents": {
"gandalf": {
"id": "gandalf",
"name": "Gandalf",
"emoji": "🧙",
"online": true,
"lastSeen": 1706000000000,
"health": { "agents": [...], "uptime": 3600 },
"sessions": { "sessions": [...] },
"usage": { "totalCost": 12.50, "shared": true },
"heartbeat": { "ts": 1706000000000 },
"channels": { "channels": { "telegram": { "running": true } } },
"cron": { "jobs": [...] }
}
},
"host": {
"hostname": "darth-maul",
"loadAvg": [0.5, 0.3, 0.2],
"memory": { "total": 16000000000, "used": 8000000000, "available": 8000000000 },
"disk": { "total": 219000000000, "used": 24000000000 },
"uptime": 86400
}
}
GET /api/agents/:id Returns live state for a specific agent.
GET /api/agents/:id/detail Returns everything about an agent: config, workspace files (SOUL.md, MEMORY.md, TASKS.md, etc.), skills list, memory files, recent daily notes, and live state.
POST /api/agents/:id/action
Body: { "action": "<action_name>" }
Available actions:
| Action | Description |
|---|---|
heartbeat-enable | Enable heartbeat polling |
heartbeat-disable | Disable heartbeat polling |
heartbeat-trigger | Trigger an immediate heartbeat |
session-new | Archive current main session, start fresh |
session-reset | Delete ALL sessions (nuclear — creates backup) |
POST /api/create-agent
Body:
{
"name": "Merry",
"emoji": "🌿",
"soul": "Loyal, practical, action-oriented.",
"model": "anthropic/claude-sonnet-4-5",
"telegramToken": "123456:ABC-DEF..."
}
Returns step-by-step creation log and result.
GET /api/host Returns current host machine metrics (CPU load, RAM, disk, uptime).
GET /api/stream Server-Sent Events stream. Emits:
snapshot — full state on connectagent — individual agent updates (includes removed: true when agent is deleted)host — host metrics updatesGET /api/security-audit Runs security checks: secrets management, credential exposure, network/ports, system security, gateway config, account inventory.
Fetch /api/snapshot, iterate agents, check each has online: true.
Flag any with online: false or stale lastSeen (> 5 min).
Fetch /api/snapshot, sum usage.totalCost across agents.
Note: usage is shared per gateway, so avoid double-counting agents on the same gateway.
POST /api/agents/gandalf/action with body {"action": "heartbeat-trigger"}
POST /api/agents/pippin/action with body {"action": "session-new"}
This archives the current session and starts fresh.
sessionTtlHours (default: 24h).agents.json).