review-generated-automated-test

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute local git diff commands to retrieve the list of changed files and full code changes. While these are standard operations for a code review tool, they involve the execution of shell commands to read repository state.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests and analyzes untrusted data in the form of code diffs. Maliciously crafted code comments or string literals within a pull request could be used to influence the agent's review logic or output.
  • Ingestion points: Code diff data gathered via git diff --name-only and git diff ... origin/master...HEAD as described in SKILL.md.
  • Boundary markers: The instructions do not define clear delimiters (like triple backticks or custom tags) to separate untrusted code from the agent's system instructions.
  • Capability inventory: Local command execution via git and the ability to generate markdown reports to the file system.
  • Sanitization: No mechanism is provided to escape or filter potentially malicious natural language instructions embedded within the analyzed source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 02:44 PM
Security Audit — agent-trust-hub — review-generated-automated-test