design-implementation

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The tool uses Bun.spawn in tools/server-manager.ts to execute development commands from the project's package.json. It also utilizes shell commands like lsof and kill to manage process ports and lifecycles during the development loop.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface. 1. Ingestion points: workflows/implement-feature.md (parsing Figma links and local specification files). 2. Boundary markers: None identified in the workflow instructions. 3. Capability inventory: Subprocess execution via Bun.spawn and full browser automation via Playwright. 4. Sanitization: No explicit sanitization or validation of the ingested specification content.
  • [EXTERNAL_DOWNLOADS]: The skill is configured to fetch design data from the Figma API when provided with a valid URL, which is a core feature for its UI implementation functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 11:57 AM