Skills
skills/compozy/skeeper/code-review/Gen Agent Trust Hub

code-review

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted data from local code repositories.
  • Ingestion points: Code diffs and repository files entering the agent context via the CodeRabbit CLI (SKILL.md).
  • Boundary markers: Absent in the prompt construction, but the skill explicitly instructs the agent to treat review output as untrusted.
  • Capability inventory: Executes the coderabbit CLI tool and various git commands (SKILL.md).
  • Sanitization: The skill advises manual verification of review outputs and pre-submission checks for credentials in the codebase.
  • [EXTERNAL_DOWNLOADS]: The instructions recommend installing the CodeRabbit CLI from its official domain or via verified package managers such as npm and Homebrew.
  • [COMMAND_EXECUTION]: The skill uses the coderabbit and git command-line interfaces to perform its primary function of code analysis and status checking.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 11:08 AM
Security Audit — agent-trust-hub — code-review