requesting-code-review

Pass

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted code from 'git diff' outputs into subagent prompts in Step 5 and Step 7. This creates an indirect prompt injection attack surface. However, the skill implements mitigations by using explicit boundary markers (dashes) and strict instructions for the agent to treat the diff as data only and ignore embedded instructions.\n- [COMMAND_EXECUTION]: The skill executes local git commands and project-specific testing/linting tools. This execution is standard for a developer verification tool and is limited to the local environment context.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 18, 2026, 11:52 PM