The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to run several command-line tools, including uv, pre-commit, mypy, pulumi, and dg. These are used according to their primary intended purpose for linting, type checking, and infrastructure previewing.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because the validation tools process project files (such as configuration files and source code) which could contain malicious instructions if the repository is untrusted.
Ingestion points: Local codebase files, including .pre-commit-config.yaml, mypy configuration, Pulumi programs, and Dagster workspace files.
Boundary markers: Not present; the instructions do not include specific delimiters or warnings to ignore instructions embedded in the files being validated.
Capability inventory: Execution of subprocesses via uv run, pulumi preview, and dg check which can run code defined in the repository.
Sanitization: No manual sanitization is performed on the files before they are processed by the validation tools.

validate-before-commit