reference-searcher

Pass

Audited by Gen Agent Trust Hub on Mar 6, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a workflow for processing untrusted data from GitHub and web searches, creating an indirect prompt injection surface. This is inherent to the skill's primary purpose of reference searching.
  • Ingestion points: External data entering via web search and GitHub code search results as described in SKILL.md.
  • Boundary markers: None specified for separating search results from agent instructions.
  • Capability inventory: Suggestions to use the GitHub CLI (gh) for read-only searches and web search tools.
  • Sanitization: No explicit sanitization or validation of the retrieved external content is mentioned.
  • [COMMAND_EXECUTION]: Includes instructional examples for using the GitHub CLI (gh) to perform code searches. These commands are limited to information gathering and do not pose a risk of system modification or dangerous execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 6, 2026, 10:21 PM