The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes local shell commands including npx vigiles generate-types, npm run build, and npx vigiles compile. These tools are used to inspect the project's enabled linting rules and ensure that code modifications do not introduce build errors.
[SAFE]: The skill reads project-specific files like .spec.ts and .vigiles/generated.d.ts, and accesses linter documentation in a sibling directory (../linter-docs/). This file access is limited to the project scope and is necessary for the skill's functionality.
[SAFE]: The instructions do not contain any network-based operations, hardcoded credentials, or attempts to bypass security restrictions. The use of disable-model-invocation: true in the frontmatter further limits the agent's execution environment.

strengthen