code-review

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute local shell commands for version control management, specifically git operations like rev-parse and log, and for verification tasks such as running test suites and build scripts.
  • [PROMPT_INJECTION]: An indirect prompt injection surface exists because the agent is instructed to identify and run verification commands based on the local codebase context. 1. Ingestion points: Local project files and identified build or test scripts. 2. Boundary markers: None explicitly defined. 3. Capability inventory: Execution of arbitrary shell commands for verification. 4. Sanitization: None provided, as it relies on the agent's logic to verify tasks within the local environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 03:54 AM
Security Audit — agent-trust-hub — code-review