context-engine-builder
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from multiple potentially untrusted sources to generate context blueprints, which could allow maliciously crafted text in source code or documentation to influence the agent's behavior during generation.\n
- Ingestion points: The skill reads local project files including .ds-ops-config.yml, component source code, documentation files, and READMEs. It also fetches metadata from Figma and Storybook APIs.\n
- Boundary markers: The instructions do not specify any delimiters or safety markers to isolate the content of the analyzed files from the generation logic.\n
- Capability inventory: The skill writes multiple files to the local .ai/context-engine/ directory and performs network requests to fetch integration data.\n
- Sanitization: No validation or sanitization of the natural language content extracted from the files is mentioned.\n- [EXTERNAL_DOWNLOADS]: The skill allows for the automated retrieval of design token definitions, component structures, and accessibility annotations from well-known services including Figma and Storybook, provided they are enabled in the project configuration.
Audit Metadata