yuandian-law-search
Pass
Audited by Gen Agent Trust Hub on Apr 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The SKILL.md file provides a bash command intended for the agent to verify the existence and content of the scripts/.env file, ensuring the API key is configured correctly before processing search requests.
- [DATA_EXFILTRATION]: The skill transmits the YD_API_KEY as a plaintext query parameter in the URL (e.g., http://aiapi.ailaw.cn:8319/search?api_key=...). This practice exposes sensitive credentials to potential logging in proxy servers, network monitoring tools, and server-side logs.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves external legal content and case summaries and incorporates them into the agent's context without any sanitization or the use of boundary markers.
- Ingestion points: Data enters the context via scripts/yd_search.py which fetches law and case data from the ailaw.cn API endpoints.
- Boundary markers: Absent. API responses are formatted into standard Markdown without delimiters to separate external content from instructions.
- Capability inventory: Subprocess execution of Python scripts, file system read access for .env configuration, and network access to the API domain.
- Sanitization: Absent. The script directly interpolates API response strings into the output stream.
Audit Metadata