rules-reviewer

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [SAFE]: Analysis of the skill instructions and supporting documents reveals no malicious patterns, unauthorized command execution, or network exfiltration attempts.
  • [NO_CODE]: The skill consists exclusively of markdown documentation and guidance templates. No executable code, scripts, or binary dependencies are included.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads and analyzes project files which may contain untrusted content. 1. Ingestion points: Reads .builderrules, .mdc, agents.md, and all .md files in the project. 2. Boundary markers: No specific delimiters or safety instructions are defined for rule ingestion. 3. Capability inventory: Standard file system search and read capabilities. 4. Sanitization: No sanitization or validation of rule content is performed prior to analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 04:45 PM
Security Audit — agent-trust-hub — rules-reviewer