The Agent Skills Directory

[SAFE]: The skill implements standard frontend architectural patterns, including Pinia stores, Vue 3 composables, and service layers with environment-based configuration for API endpoints. It follows industry best practices for error handling and data mapping.
[COMMAND_EXECUTION]: The skill uses terminal tools to execute standard development workflows, such as npx vue-tsc --noEmit for type verification and npm run build for verifying project builds. These operations are restricted to the local development environment and are appropriate for a code-generation agent.
[PROMPT_INJECTION]: The skill processes external specification files (enriched-spec.md, spec.md) to drive code generation, creating a surface for indirect prompt injection.
Ingestion points: Specification files identified in SKILL.md Step 0.
Boundary markers: None identified in the provided instructions to differentiate between agent instructions and data input.
Capability inventory: File system writes via mcp-fs and command execution via mcp-terminal across all logic implementation steps.
Sanitization: No explicit sanitization or validation of the input specification content is performed.
Note: This behavior is typical for development agents and is managed through safe operational practices.

logic-coder