The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands for various build systems (npm, yarn, pnpm, pip, poetry, gradle, maven, cargo, go, make) to build and test code based on detected environment files.
[DYNAMIC_EXECUTION]: The skill facilitates dynamic command generation and execution through user-provided flags (--cmd, --filter) and configuration overrides defined in the project's .claude/config.json file.
[EXTERNAL_DOWNLOADS]: Execution of standard build commands such as pip install -r requirements.txt and poetry install involves downloading packages from external package registries.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted local project files to determine which shell commands to execute, creating a potential attack surface if those files are maliciously crafted.
Ingestion points: Project manifest and lock files including package.json, yarn.lock, pnpm-lock.yaml, pyproject.toml, requirements.txt, Cargo.toml, go.mod, build.gradle, pom.xml, Makefile, and the configuration file .claude/config.json.
Boundary markers: Absent.
Capability inventory: Ability to execute arbitrary shell commands for build and test workflows.
Sanitization: Absent.

build-system