gsd-spec-phase

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute localized workflow steps defined within the gsd-core framework.
  • [DATA_EXPOSURE]: It reads local project documentation, such as PROJECT.md and REQUIREMENTS.md, to establish the necessary context for requirement clarification.
  • [PROMPT_INJECTION]: The 'codebase scouting' feature involves reading arbitrary files from the project, which constitutes an indirect prompt injection surface typical of development-oriented AI tools.
  • Ingestion points: Reads files from the active codebase and specific documentation (e.g., ROADMAP.md, STATE.md).
  • Boundary markers: None explicitly defined in the skill body.
  • Capability inventory: Access to Bash, Write, and Read tools.
  • Sanitization: Not explicitly defined in this layer; relies on underlying agent guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 09:38 PM
Security Audit — agent-trust-hub — gsd-spec-phase