component
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill triggers shell commands for quality gates such as linting and typechecking during the component development loop and uses grep for codebase auditing, as defined in SKILL.md and references/actions/dev.md.
- [PROMPT_INJECTION]: An indirect prompt injection surface exists as the skill ingests external design context from Figma links and uses it to drive code generation and quality checks.
- Ingestion points: Figma design context extraction in references/actions/structure.md and references/actions/dev.md via the get_design_context Figma MCP tool.
- Boundary markers: Absent; the skill does not implement specific delimiters or instructions to ignore embedded prompts within the ingested design data.
- Capability inventory: File write access (specs, components, and reports) and shell command execution (linting and typechecking) as documented in SKILL.md.
- Sanitization: No explicit sanitization or validation of strings extracted from Figma design context before being used in prompt interpolation or file generation.
Audit Metadata