create-plan

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Git commands (git log, git diff) to analyze project history and codebase state during the research phase.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests untrusted data from the codebase and GitHub issues which could contain malicious instructions.
  • Ingestion points: Codebase files, existing GitHub issues, and project comments (Step 0).
  • Boundary markers: No explicit delimiters are used to separate ingested data from agent instructions.
  • Capability inventory: Execution of shell commands (Git) and writing to GitHub issues via the #issue_write capability.
  • Sanitization: The instructions do not specify any sanitization or validation of the data retrieved from external sources before processing or outputting it.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 05:28 PM