The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a wide range of shell commands to perform linting, type checking, testing, and building. This includes running scripts defined in the project's package.json (e.g., npm run lint, npm run build), searching for files using find and grep, and checking git diffs for sensitive information.
[REMOTE_CODE_EXECUTION]: The skill utilizes npx to execute various well-known development utilities such as eslint, biome, tsc, vitest, and jest. These operations involve fetching and running code from the public npm registry.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of development dependencies, specifically suggesting and executing the installation of husky and lint-staged via npm install. These are well-known services commonly used for managing git hooks.
[PERSISTENCE]: The skill explicitly configures git hooks using Husky (e.g., npx husky init and writing to .husky/pre-commit). This establishes a persistence mechanism where the quality gate logic is automatically executed on every code commit.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes data from the local repository.
Ingestion points: The skill reads package.json, tsconfig.json, eslintrc*, and biome.json files, and it parses the output of git diff and various test/linter logs.
Boundary markers: There are no explicit delimiters or warnings to ignore instructions embedded within the files being analyzed.
Capability inventory: The skill has the ability to execute shell commands (npm, npx, find, git) and perform file operations (rm -rf for build cleanup, Husky configuration).
Sanitization: The skill does not implement sanitization or filtering of content extracted from repository files or tool outputs before it is processed.

quality-gates