Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Feishu Repair
v2.0.0自动修复飞书-自动修复飞书群聊+自动修复会话 - 诊断 Gateway 连接、权限配置、消息投递问题
⭐ 0· 47·0 current·0 all-time
byc32@amd5
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's stated purpose (diagnose and repair OpenClaw Feishu channel) matches the code: it reads ~/.openclaw/openclaw.json backups, sets channels.feishu.* via the openclaw CLI, restarts the openclaw-gateway.service, and calls Feishu APIs. However, the registry metadata declares no required binaries or env vars while the script actually depends on systemctl/journalctl, the openclaw CLI, curl, and a POSIX sleep — the missing declared dependencies is an inconsistency.
Instruction Scope
SKILL.md and the script instruct the agent to read the user's ~/.openclaw config and backups, edit live OpenClaw config via 'openclaw config set', force-restart the Gateway service, inspect journalctl logs, and send validation messages to all configured group/user targets. These actions are within the stated repair scope, but they involve reading sensitive credentials and proactively messaging all contacts — behaviour that impacts privacy and production systems and should be explicitly consented to.
Install Mechanism
There is no install spec (instruction-only plus an included script). No remote downloads or archive extraction occur during install. The runtime script will be executed by the agent when invoked; that is expected for an instruction-only skill.
Credentials
The skill does not declare environment variables but reads appId/appSecret and allowFrom/groupAllowFrom from local openclaw.json files. Accessing those credentials is necessary to obtain Feishu tenant_access_token and send validation messages, so the access is functionally justified — but these are sensitive secrets and the skill will transmit derived tokens to Feishu APIs and send messages on behalf of the app.
Persistence & Privilege
The skill will modify live OpenClaw configuration via 'openclaw config set' and will forcibly restart the user-level openclaw-gateway.service without further confirmation. While 'always' is false and it does not auto-enable itself, these are high-privilege actions with real operational impact; users should be aware it will change config and send messages to all targets.
What to consider before installing
What to consider before installing or running this skill:
- It will read ~/.openclaw/openclaw.json and backup files (which contain appId/appSecret and access lists). Those app secrets are sensitive — inspect the script and confirm you trust it before allowing access.
- The script requires externally available tools that are not declared in metadata: openclaw CLI, systemctl/journalctl (systemd user service), curl, and sleep. Ensure these are present and that running them is acceptable in your environment.
- The skill will modify live configuration (using 'openclaw config set') and forcibly restart the openclaw-gateway.service automatically. Back up current config and test in a non-production environment first.
- It will iterate over allowFrom and groupAllowFrom and send messages to all listed chats/users to validate repairs. Consider privacy and inform impacted users/groups before running.
- If you want to reduce risk: run the diagnose script manually in a dry-run or JSON-only mode first, review the output, and only allow fixes after confirming the proposed changes; or edit the script to add an explicit interactive confirmation step before any 'openclaw config set' or 'systemctl restart' call.
- If you are unsure, ask the skill author for provenance or run the script in an isolated environment where it cannot reach the real Feishu API or restart production services.scripts/diagnose.js:92
Shell command execution detected (child_process).
Patterns worth reviewing
These patterns may indicate risky behavior. Check the VirusTotal and OpenClaw results above for context-aware analysis before installing.Like a lobster shell, security has layers — review code before you run it.
latestvk97a1ydhh41rc6m1t5wfe95y0s84v97m
License
MIT-0
Free to use, modify, and redistribute. No attribution required.