copilot-init
Pass
Audited by Gen Agent Trust Hub on Jun 14, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill audits untrusted repository content to generate instructions and workflows, creating a surface for indirect prompt injection where malicious files could influence the agent's output.\n
- Ingestion points: Uses
Read,Glob, andGrepto audit repository files and toolchains.\n - Boundary markers: Does not implement delimiters or ignore markers for processed repository data.\n
- Capability inventory: Employs
Write,Edit, andBashto create executable GitHub workflows.\n - Sanitization: Lacks explicit requirements for sanitizing repository data before inclusion in generated files.\n- [COMMAND_EXECUTION]: The skill uses
Bashto audit the environment and validate setup steps, which is necessary for its function but provides a mechanism for executing commands derived from the repository auditing phase.
Audit Metadata