doc-tspec-audit
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill defines a benign orchestration workflow for documentation quality gates. The execution sequence (validator, reviewer, report generation) is a standard administrative task without suspicious commands or unauthorized network activity.- [NO_CODE]: The primary skill file contains only a filesystem path reference rather than executable code or scripts, indicating it likely points to an external or environment-provided implementation.- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes external markdown documentation that could contain malicious instructions. However, the use of a formal validator gate and deduction-based scoring mitigates this risk. 1. Ingestion points: TSPEC documentation files at
docs/10_TSPEC/. 2. Boundary markers: None explicitly defined in the orchestration instructions. 3. Capability inventory: Orchestrates sub-skills (doc-tspec-validator,doc-tspec-reviewer) and writes audit reports to the filesystem. 4. Sanitization: Relies on structural schema validation and semantic review thresholds.
Audit Metadata