research-git

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from untrusted public GitHub repositories, creating a surface for indirect prompt injection attacks. If a scanned repository contains malicious instructions, the agent could theoretically be influenced. However, the skill provides strong mitigation through its defined workflow, which requires the agent to rewrite insights in its own voice, use a manual research pack review process before any merging, and explicitly treat all fetched files as untrusted data that should never be executed.
  • [EXTERNAL_DOWNLOADS]: The skill uses the gh CLI and curl to fetch repository assets and security metadata from GitHub and the OpenSSF Scorecard API. These operations are essential to the skill's primary function and target well-known, reputable services. The documentation includes clear instructions on managing API rate limits and verifying source integrity.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 02:03 PM