full-repo-review

Pass

Audited by Gen Agent Trust Hub on Jun 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses find to identify source files and executes a local Python utility at ~/.claude/scripts/score-component.py to perform deterministic health checks. This is standard behavior for repository analysis tools.
  • [PROMPT_INJECTION]: The skill analyzes arbitrary source code across the repository, creating an indirect prompt injection surface inherent to its function as a code reviewer.
  • Ingestion points: Scans and reads files from scripts/, hooks/, skills/, agents/, and docs/ directories (SKILL.md, Phase 1).
  • Boundary markers: None are explicitly defined in the orchestration logic; the skill relies on the downstream review engine for input handling.
  • Capability inventory: The skill has Bash for local scanning and Write to generate the final report file.
  • Sanitization: No sanitization is performed on file contents, as the tool is intended to analyze the original source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 20, 2026, 02:02 PM
Security Audit — agent-trust-hub — full-repo-review