tech-spec

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a mechanical documentation generator, ensuring that technical views are strictly derived from source code without human or AI-authored business content, thereby maintaining a single source of truth.\n- [COMMAND_EXECUTION]: The skill invokes an internal Node.js script (scripts/generate-tech-specs.mjs) to automate the documentation process. The script is bundled within the skill's directory and its file system operations are restricted to the repository root.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes source code and test annotations (ingestion surface in scripts/generate-tech-specs.mjs). This potential attack vector is mitigated by the mechanical nature of the generator (the C8 contract), which uses fixed templates and explicitly forbids interpreting, judging, or narrating the ingested data. Evidence: 1. Ingestion points: Source code files and test annotations scanned by the internal script. 2. Boundary markers: Every generated file must carry a mandatory > DERIVED banner. 3. Capability inventory: File system read/write/delete operations restricted to the project root. 4. Sanitization: Use of fixed templates and strict verification of derived banners to prevent content injection.\n- [DYNAMIC_EXECUTION]: The internal script manages the technical spec tree using Node.js file system APIs. It incorporates safety features including symbolic link rejection, transaction-based reconciliation with rollback capabilities, and verification that only previously derived markdown files are modified or deleted.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 10:12 AM