qinyan-topic-analysis

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a shell script (scripts/search.sh) to execute curl commands for literature retrieval. This is a primary function of the skill and is constrained to specific academic API endpoints.
  • [EXTERNAL_DOWNLOADS]: The skill performs POST requests to api.qinyanai.com to fetch research paper metadata. These requests are authenticated via a user-provided QINYAN_API_KEY and target the official vendor infrastructure.
  • [CREDENTIALS_UNSAFE]: The documentation correctly instructs users to manage their API keys via environment variables (QINYAN_API_KEY) rather than hardcoding them into the scripts, which is a recommended security practice.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external research data (titles, abstracts, and metadata) from Google Scholar, ArXiv, and PubMed. While this is an ingestion surface for untrusted data, the risk is low as the data is used for academic summarization and topic analysis rather than being executed or used in sensitive logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 08:21 AM
Security Audit — agent-trust-hub — qinyan-topic-analysis