happycapy-feishu

Warn

Audited by Gen Agent Trust Hub on Mar 21, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Modifies the global agent configuration file ~/.claude.json using a Python script to register the MCP server locally.
  • [COMMAND_EXECUTION]: Performs in-place source code patching of the @larksuiteoapi/lark-mcp package using perl to increase hardcoded authentication timeout values.
  • [COMMAND_EXECUTION]: Overwrites the keytar dependency within the installed Node.js module with a custom script (keytar-file-storage.js) to provide file-based credential storage.
  • [EXTERNAL_DOWNLOADS]: Fetches the official @larksuiteoapi/lark-mcp package from the NPM registry using the npx utility.
  • [REMOTE_CODE_EXECUTION]: Executes code downloaded from the NPM registry via npx to facilitate the OAuth login process and run the MCP service.
  • [PROMPT_INJECTION]: The skill creates a significant attack surface for indirect prompt injection. 1. Ingestion points: Callback URLs provided by users and external Feishu content (messages, documents, and Bitable records). 2. Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the processing logic. 3. Capability inventory: The skill has broad write capabilities including replying to messages, updating records, and writing to documents via the registered MCP tools. 4. Sanitization: No sanitization or validation of the ingested external content is performed before processing.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 21, 2026, 07:18 AM