caveman-review

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted code diffs from pull requests. Malicious code within a PR could potentially contain instructions aimed at influencing the AI agent's behavior during the review process.
  • Ingestion points: The skill ingests pull request diffs and code contents provided in the agent context.
  • Boundary markers: There are no explicit delimiters or boundary instructions provided to the agent to distinguish between the skill instructions and the potentially malicious data within the diffs.
  • Capability inventory: The skill is limited to generating textual comments for PR reviews. It does not list any allowed-tools in its frontmatter and does not demonstrate capabilities for file writing, network operations, or subprocess execution in its instructions.
  • Sanitization: The instructions do not include any logic for sanitizing or escaping the content of the diffs before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 01:59 AM
Security Audit — agent-trust-hub — caveman-review