Skillv1.0.0

ClawScan security

Feishu Multi-Bot 飞书多机器人配置 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 13, 2026, 5:32 AM

Verdict: Benign
Confidence: high
Model: gpt-5-mini
Summary: Instructions and requirements are coherent with the stated purpose (configuring multiple Feishu bots for OpenClaw agents); no unrelated privileges, installs, or external endpoints are requested.
Guidance: This skill appears to do what it says, but take these precautions before using it: (1) Only provide App ID/App Secret values to the agent in a private, trusted channel; do not paste them in public logs or channels. (2) Consider whether dmPolicy: "open" is appropriate — it allows anyone who adds the bot to chat without pairing. If unsure, keep pairing enabled. (3) Back up your current Gateway configuration before applying config.patch and be prepared for a restart (possible temporary downtime). (4) After applying, verify using openclaw gateway status and openclaw logs --follow; rotate App Secrets if you suspect they were exposed. (5) Ask the skill/agent to avoid echoing secrets in visible outputs or logs; if the SKILL.md is to be used as an automated instruction, add an explicit step to redact secrets from any agent responses or logs.

Review Dimensions

Purpose & Capability: okThe skill's name/description (multi-bot Feishu configuration) matches the runtime instructions: it collects per-agent Feishu App ID/App Secret pairs and writes them into the OpenClaw Gateway channels.feishu.accounts configuration via gateway config.patch. Nothing requested or described is unrelated to that purpose.
Instruction Scope: noteInstructions stay within expected scope (gather credentials from user, build JSON config, call gateway config.patch, check gateway status/logs). Two points to note: (1) the guide suggests setting dmPolicy:"open" (disables pairing), which is a security/privacy decision the administrator should consider before enabling; (2) the SKILL.md warns to keep App Secrets confidential but does not explicitly instruct the agent how to handle/display these secrets safely (for example, avoid echoing them in logs or chat).
Install Mechanism: okInstruction-only skill with no install spec and no code files — nothing will be downloaded or written by an install step.
Credentials: okNo environment variables or unrelated credentials are requested. The only sensitive inputs are the per-account App ID and App Secret which are necessary and proportionate for configuring Feishu accounts.
Persistence & Privilege: okSkill does not request always:true or elevated platform privileges. The operational action (config.patch) legitimately modifies the Gateway's channel configuration and may trigger a restart as documented — this is consistent with its purpose and limited in scope.