skills/atxinsky/skills/feishu-bridge/Gen Agent Trust Hub

feishu-bridge

Warn

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [PERSISTENCE_MECHANISMS]: The skill instructions configure a macOS LaunchAgent to ensure the bridge service starts automatically and restarts if it fails, which is a persistence mechanism. Evidence: launchctl load command for com.clawdbot.feishu-bridge.plist in SKILL.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill bridges external user messages from Feishu to the AI agent, providing a vector for untrusted data to influence agent behavior. (1) Ingestion points: Feishu WebSocket event im.message.receive_v1. (2) Boundary markers: None identified in instructions. (3) Capability inventory: The agent processes and replies to group and P2P messages based on user triggers. (4) Sanitization: No sanitization or filtering logic is specified for incoming message content.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill manages sensitive credentials and configuration files within the user's home directory. Evidence: Mentions of ~/.clawdbot/secrets/feishu_app_secret and ~/.clawdbot/clawdbot.json as required environment variables and storage locations.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 14, 2026, 04:00 PM
Security Audit — agent-trust-hub — feishu-bridge