tailrocks-grpc-review

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external code and diffs, which are untrusted ingestion points.
  • Ingestion points: Defined in SKILL.md as 'gRPC diff, module, or whole service surface'.
  • Boundary markers: references/runtime-trust.md provides explicit instructions to treat repository files as untrusted data and ignore any instructions embedded within them.
  • Capability inventory: SKILL.md allows the agent to execute repository code for verification purposes (e.g., wire tests) using cargo or buf.
  • Sanitization: The skill implements strict execution constraints, including network disablement, secret scrubbing, and hashing of repository state to prevent mutation.
  • [COMMAND_EXECUTION]: The skill instructions permit the execution of shell commands (e.g., buf lint, buf breaking, cargo test) to verify contract compliance.
  • Evidence: SKILL.md section 5 ('Use commands only under explicit authority') outlines strict controls for using commands, including active-task authority, disabled network, and frozen inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:13 PM
Security Audit — agent-trust-hub — tailrocks-grpc-review