github-copilot-starter

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the fetch tool to retrieve instruction templates, agent definitions, and skill presets from the official GitHub 'awesome-copilot' repository. This practice ensures the generated configuration adheres to current industry standards for GitHub Copilot.
  • [PROMPT_INJECTION]: The skill has an architectural surface for indirect prompt injection because it ingests external data to populate local configuration files and GitHub Action workflows.
  • Ingestion points: Content retrieved via the fetch tool from the https://github.com/github/awesome-copilot repository.
  • Boundary markers: None; the skill does not implement delimiters or 'ignore' instructions to isolate the fetched external content from the agent's logic during file creation.
  • Capability inventory: The skill possesses file-write capabilities across multiple sensitive directories including .github/instructions/, .github/agents/, and .github/workflows/.
  • Sanitization: There is no evidence of validation or sanitization of the fetched remote content before it is integrated into the local project structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 07:50 AM
Security Audit — agent-trust-hub — github-copilot-starter