migrate-agents-to-copilot-instructions

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates the migration of instruction content from potentially untrusted or unverified source files (such as AGENTS.md, CLAUDE.md, or GEMINI.md) into core GitHub Copilot configuration files. This creates a surface for indirect prompt injection where malicious instructions could be mirrored into the repository's permanent settings.\n
  • Ingestion points: Content is ingested from root and nested AGENTS.md, CLAUDE.md, and GEMINI.md files (SKILL.md).\n
  • Boundary markers: The skill lacks explicit instructions for the agent to use delimiters or "ignore embedded instructions" markers when writing the new Copilot configuration files.\n
  • Capability inventory: The skill has the capability to read any file in the repository and write to the .github directory (SKILL.md).\n
  • Sanitization: There is no requirement or instruction for the agent to sanitize, validate, or filter the source instructions during the transformation process (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:29 AM