research-lookup

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to the OpenRouter API (openrouter.ai) to retrieve research summaries and citations. This is the primary intended functionality of the tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests text from an external API (OpenRouter), which constitutes an untrusted data source. This is a common architectural characteristic of research tools.
  • Ingestion points: Data enters the context via API response content in research_lookup.py.
  • Boundary markers: The output from the API is integrated into the agent's response without specific delimiters or isolation instructions.
  • Capability inventory: The skill frontmatter allows the use of Bash, Read, Write, and Edit tools.
  • Sanitization: The skill identifies DOIs and URLs but does not sanitize the narrative summary text returned by the model.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 02:49 PM