external-reference-research

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill focuses on information retrieval and synthesis, providing structured workflows for developers to research libraries and frameworks. It emphasizes best practices such as prioritizing official documentation and cross-referencing information.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection due to its core function of processing external, untrusted data.
  • Ingestion points: The agent retrieves content from public GitHub repositories and general web search results (SKILL.md).
  • Boundary markers: The instructions do not define clear delimiters or system instructions to ignore embedded commands within the data fetched from the web or GitHub.
  • Capability inventory: The skill utilizes GitHub CLI, Web Search, and Context7 to browse and retrieve external content into the agent's context.
  • Sanitization: There are no instructions for the agent to sanitize, escape, or validate the content retrieved from external sources before it is analyzed or presented to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 08:31 AM
Security Audit — agent-trust-hub — external-reference-research