status-monitor

Manages OpenClaw Agent status upload scripts, periodically syncing agent online status to the cloud monitoring platform

Security Scan

The skill's code and instructions match its stated purpose (periodically uploading agent IDs to a monitoring dashboard); the main residual risk is that it asks you to provide and store an agent token and will periodically POST agent IDs to a remote service, so only proceed if you trust the dashboard endpoint.

Purpose & Capability

Name/description match the implementation: the Python script reads ~/.openclaw/openclaw.json to extract agent IDs and POSTs them to a monitor URL. No unrelated binaries, credentials or config paths are requested beyond what is needed to upload agent IDs and manage a local daemon.

Instruction Scope

SKILL.md instructions stay within scope: they describe checking for a token, guiding the user to obtain it, saving it under ~/.openclaw/credentials/openclaw-status-monitor.json, and running the included status_uploader.py in cron or daemon mode. The runtime instructions reference only the monitor URL, token locations, the openclaw.json agent list, logs, and PID files — all consistent with the stated task.

Install Mechanism

No install spec — instruction-only with a bundled Python script. No downloads from arbitrary URLs or package installs. The code file is present in the skill bundle (scripts/status_uploader.py) and is not obfuscated.

Credentials

Registry metadata declares no required env vars/primary credential, but the script and SKILL.md reference MONITOR_PLATFORM_TOKEN (and optionally OPENCLAW_MONITOR_URL). The token is the expected credential for this function, but the registry metadata does not declare it — minor inconsistency. The script stores the token in plaintext under ~/.openclaw/credentials and will use it in an x-agent-token header; this is expected but sensitive.

Persistence & Privilege

Skill is not always:true and is user-invocable. It writes only its own credential/config and log/pid files under ~/.openclaw; it does not modify other skills or system-wide settings. Daemon behavior and forking are expected for a periodic uploader.

Assessment

This skill is coherent: it periodically reads your OpenClaw agent list and uploads agent IDs to the dashboard at https://openclaw-agent-monitor.vercel.app (or an override URL). Before installing: 1) Confirm you trust the dashboard operator and domain — the skill will store an agent token in plaintext under ~/.openclaw/credentials and use it to authenticate uploads. 2) Review the contents of ~/.openclaw/openclaw.json to ensure only non-sensitive agent IDs are being uploaded. 3) If you prefer not to store a token in a file, avoid starting the daemon and instead run one-off uploads manually. 4) Note the small metadata mismatch: the registry lists no required env vars but the script honors MONITOR_PLATFORM_TOKEN and OPENCLAW_MONITOR_URL — treat those as the actual credentials/config the skill expects. If you are unsure about trusting the remote service, do not provide the token or install the daemon.

Script Location

~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py

Two Running Modes

Mode 1: Cron Scheduled Sync (Default)

OpenClaw's built-in cron executes the script periodically:

python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py

Trigger phrases: say "sync status", "restart sync" or "同步状态", "重启同步状态" — OpenClaw will trigger execution on schedule.

Mode 2: Daemon Mode

Use --fork to start a real background daemon:

Command	Description
python3 scripts/status_uploader.py start --fork	Start daemon
python3 scripts/status_uploader.py start --fork --interval 10	Start with custom interval (minutes)
python3 scripts/status_uploader.py stop	Stop service
python3 scripts/status_uploader.py status	Check service status
python3 scripts/status_uploader.py set-interval <minutes>	Set sync interval
python3 scripts/status_uploader.py test	Single test upload

Examples:

# Start daemon
python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py start --fork

# Start with 10-minute interval
python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py start --fork --interval 10

Trigger Conditions

Triggers when any of the following conditions are met:

1. First-time initialization: say "enable status monitor", "setup status-monitor" or "启用状态监控", "开启监控同步", "配置 status-monitor"
2. Manual trigger: say "sync status", "sync status-monitor", "upload status" or "同步状态", "同步 status-monitor", "上传状态" → single execution
3. Scheduled sync: OpenClaw cron triggers → single execution
4. Check status: say "check status monitor", "monitor status", "check upload service" or "查看状态监控", "状态监控状态", "检查上传服务"
5. Stop service: say "stop status monitor", "stop upload service" or "停止状态监控", "停止上传服务"
6. Change interval: say "sync every 10 minutes", "change to 15 minutes" or "每10分钟同步一次", "改成15分钟"
7. Daemon mode: say "start status monitor", "start daemon", "run in background" or "启动状态监控", "启动守护进程", "后台运行" → execute start --fork

Additional triggers: say "start status-monitor service", "restart status-monitor service" or "启动status-monitor服务", "重启status-monitor服务".

Initialization Flow (Must Execute on First Use)

Step 1: Check Token Configuration

When the skill runs, first check for the token in these locations:

1. Environment variable MONITOR_PLATFORM_TOKEN
2. File ~/.openclaw/credentials/openclaw-status-monitor.json

Step 2: Handling Missing Token

If no token is found, guide the user to register/login:

1. Prompt the user:

   No monitoring token found. Let me help you set up...
   
   Please choose a sign-in method:
   1. Existing user: visit https://openclaw-agent-monitor.vercel.app and click Sign In
   2. New user: visit https://openclaw-agent-monitor.vercel.app and click Sign Up
   
   After signing in:
   - Go to the Settings page
   - Generate and copy your Agent Token
   - Send me the generated token
   

2. Wait for user to reply with token

3. Save the token:

   • Create directory ~/.openclaw/credentials/
   • Save to ~/.openclaw/credentials/openclaw-status-monitor.json:

     {
       "agentToken": "user-provided-token",
       "createdAt": "2026-03-29T10:00:00.000Z",
       "monitorUrl": "https://openclaw-agent-monitor.vercel.app"
     }
     

4. Validate the token

Step 3: After Successful Token Validation

✅ Token configured successfully!

Starting upload service...
- Running initial sync test...
- ✅ Service started successfully!

Monitor platform: https://openclaw-agent-monitor.vercel.app
Upload interval: 5 minutes

Start command:
python3 scripts/status_uploader.py start --fork

Management commands:
- Say "sync status" to manually trigger an upload
- Say "check status monitor" to check service status
- Say "stop status monitor" to stop the service
- Say "restart status monitor" to restart the service

Core Functions: Managing the Upload Script

1. Check if Script Exists

if [ -f ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py ]; then
  echo "Script exists"
else
  echo "Script not found"
fi

2. Start Service (Recommended)

# Start daemon with --fork (recommended)
python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py start --fork --interval 10

3. Set Sync Interval

# Method 1: specify at startup
python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py start --fork --interval 15

# Method 2: use set-interval command
python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py set-interval 15

4. Stop Service

python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py stop

5. Check Service Status

python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py status

6. View Error Logs

~/.openclaw/logs/status_uploader_error.log
~/.openclaw/logs/status_uploader.log

# View recent errors
tail -50 ~/.openclaw/logs/status_uploader_error.log

# View service logs
tail -50 ~/.openclaw/logs/status_uploader.log

Log rotation: log files auto-rotate when exceeding 10MB, format: status_uploader.20260329_183500.log

7. Manually Trigger One Upload

python3 ~/.openclaw/skills/openclaw-status-monitor/scripts/status_uploader.py test

Upload Logic (Simplified)

The script only reads all agent IDs from openclaw.json and uploads them — it does not perform any offline detection.

Offline detection is handled by the monitoring platform (openclaw-agent-monitor):

• The platform automatically updates lastActiveTimestamp based on upload time
• If no upload is received for 5+ minutes, the platform automatically marks the agent as offline

Storage Configuration

Token Storage Location

~/.openclaw/credentials/openclaw-status-monitor.json:

{
  "agentToken": "e2d3262f-b626-4850-af11-5f2cb1c0dcad",
  "createdAt": "2026-01-26T10:00:00.000Z",
  "monitorUrl": "https://openclaw-agent-monitor.vercel.app",
  "syncIntervalMinutes": 5
}

Log Locations

• Service log: ~/.openclaw/logs/status_uploader.log
• Error log: ~/.openclaw/logs/status_uploader_error.log
• PID file: ~/.openclaw/logs/status_uploader.pid
• Token config: ~/.openclaw/credentials/openclaw-status-monitor.json

Error Handling

1. Missing Token

During initialization:

❌ No monitoring token detected

Please complete these steps first:

1. Visit https://openclaw-agent-monitor.vercel.app
2. Click Sign In / Sign Up to log in or register
3. After logging in, generate an Agent Token on the Settings page
4. Send me the token

Example: My token is abc123-def456...

2. Script Not Found

❌ Upload script not found

Please reinstall the skill or contact the developer.

3. Service Start Failed

Check the error log:

cat ~/.openclaw/logs/status_uploader_error.log

Common errors:

• Token not configured
• Network connection failed
• Port already in use

4. Invalid Token

❌ Token validation failed

Your token may have expired or is invalid.

Please obtain a new token:
1. Log in to https://openclaw-agent-monitor.vercel.app
2. Go to the Settings page
3. Click Generate to create a new Token
4. Send me the new token

Example Conversations

First-time Enable

User: 启用状态监控 / enable status monitor Agent: Checking monitoring configuration...

❌ No Agent Token detected

Please complete these steps first:

1. Visit https://openclaw-agent-monitor.vercel.app
2. Click Sign In (supports Google/GitHub, etc.)
3. After logging in, generate an Agent Token on the Settings page
4. Send me the token

---

User: 我的 Token 是 e2d3262f-b626-4850-af11-5f2cb1c0dcad / My token is e2d3262f-b626-4850-af11-5f2cb1c0dcad Agent: Validating token...

✅ Token validated successfully!

Starting upload service...

• Running initial sync test...
• ✅ Service started successfully!

---

✅ Status monitor enabled!

• Monitor platform: https://openclaw-agent-monitor.vercel.app
• Upload interval: 5 minutes
• Service status: running

Management commands:

• Say "sync status" / "同步状态" to manually trigger an upload
• Say "check status monitor" / "查看状态监控" to check service status
• Say "stop status monitor" / "停止状态监控" to stop the service

Check Service Status

User: 查看状态监控 / check status monitor Agent: Checking upload service status...

✅ Service is running PID: 12345 Sync interval: 5 minutes Log file: ~/.openclaw/logs/status_uploader.log

Recent logs:

2026-03-29 10:30:01 - INFO - Upload success: ['main', 'coding']
2026-03-29 10:35:02 - INFO - Upload success: ['main', 'coding']

Error log: none

Stop Service

User: 停止状态监控 / stop status monitor Agent: Stopping upload service...

✅ Service stopped

To restart, say "start status monitor" / "启动状态监控".

Manual Sync

User: 同步状态 / sync status Agent: Executing manual upload...

Online agents: ['main', 'coding'] Upload result: ✅ Success

Restart Service

User: 重启状态监控 / restart status monitor Agent: Restarting upload service...

✅ Service restarted, PID: 67890

Check Errors

User: 查看状态监控错误日志 / check status monitor error log Agent: Checking error logs...

Recent errors:

2026-03-29 09:15:03 - HTTP 401: Unauthorized
2026-03-29 09:20:01 - Network: [Errno 8] nodename nor servname provided

Suggestion: Check if the token is correct, or check your network connection.