spec-driven-development
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references technical guidelines and standards from GitHub's official documentation and the IEEE standards website. These references are used for methodology grounding and do not involve executable code downloads.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external project artifacts like
spec.md,plan.md, andtasks.md. This represents a surface for indirect prompt injection if those files contain untrusted or malicious instructions. - Ingestion points: Project documentation files (
spec.md,plan.md,tasks.md) read by the agent. - Boundary markers: The workflow defines clear separation between artifacts (Requirements vs. Plan) and uses phase gates to delimit work.
- Capability inventory: The skill uses
Read,Grep, andBashtools to analyze files and implement specified changes. - Sanitization: The methodology explicitly mandates "human approval when available" as a review gate before proceeding to implementation.
Audit Metadata