maestro-merge

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes content from state.json and the scratch directory, which could contain instructions if the project files are untrusted.
  • Ingestion points: Reads from state.json and files in the scratch worktree directory.
  • Boundary markers: No explicit delimiters are specified for the artifact data processing.
  • Capability inventory: Uses Bash, Write, and Edit tools to perform file operations and command execution.
  • Sanitization: The instructions do not define sanitization logic for the data ingested from milestone artifacts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 12:16 PM