dev-conventions
Pass
Audited by Gen Agent Trust Hub on Jun 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill documents legitimate coding conventions and project management workflows for an Astro-based website.
- [COMMAND_EXECUTION]: Instructions include standard development commands using
pnpmandwranglerfor linting, building, and database migrations. These are appropriate for the project context and follow established best practices for the Cloudflare D1 ecosystem. - [PROMPT_INJECTION]: The skill defines an attack surface for indirect prompt injection as it processes user requests for code generation (e.g., "add a component"). Ingestion occurs via user-supplied task descriptions in the SKILL.md metadata. While explicit boundary markers and sanitization logic are absent in the provided references, the instructions serve to constrain agent behavior to safe, project-standard patterns. Capability inventory includes standard development tasks like file creation and linting via
pnpm.
Audit Metadata