The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes project source code and coverage reports, which represents an attack surface for indirect instructions.
Ingestion points: The skill reads project metadata (package.json), configuration files (vitest.config.ts), and all source files under the src/ directory in SKILL.md (Steps 1, 2, 4, 5, and 6).
Boundary markers: The skill does not use specific delimiters or instructions to the agent to ignore potentially malicious content within the analyzed source files.
Capability inventory: The skill has high privileges, utilizing the Shell tool for package installation and test execution, and the Write tool for modifying project source code and configuration files.
Sanitization: There is no evidence of sanitization or validation of the content read from source files before it is used to generate new tests or refactor code.
[COMMAND_EXECUTION]: The skill performs various shell operations to manage the testing environment.
It uses pnpm add -D to install @vitest/coverage-v8 and other missing dependencies identified during test runs.
It executes test runners like vitest and jest via npx and npm and uses a node -e script to parse local coverage summaries.
[EXTERNAL_DOWNLOADS]: The skill downloads standard development tools and libraries.
It installs @vitest/coverage-v8 and potentially other test-related packages from the NPM registry to meet coverage requirements.

test-coverage