feishu-doc-fetch
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automatically manages a background daemon process (
daemon.mjs) using thenode:child_processspawnmethod to bridge communication between the CLI and a browser extension. This is an intended architectural component for its functionality.- [PROMPT_INJECTION]: As the skill ingests external document content from Feishu, it presents an indirect prompt injection surface where instructions embedded in a document could potentially influence agent behavior. - Ingestion points: Document content is received as data URLs from the browser extension and processed in
fetch.mjsandunzip.mjs. - Capability inventory: The skill has permissions to create directories and write files to the local filesystem.
- Boundary markers: No explicit boundary markers or "ignore instructions" warnings are added to the fetched content.
- Sanitization: The skill implements path traversal protection in
unzip.mjsby filtering out parent directory references in file paths.
Audit Metadata