← ClaudeAtlas

validatelisted

Normalize a repo Component, run its lint and test validation checks, and report each result. Use when the user asks to validate changes, lint/format code, run tests, or verify before committing or pushing.
compforge/devloop · ★ 4 · Code & Development · score 64
Install: claude install-skill compforge/devloop
For complete Component validation, run: ``` <PLUGIN_ROOT>/scripts/python <PLUGIN_ROOT>/scripts/run_validate.py [<repo name|path>] ``` `run_lint.py` uses changed-file scope when the project explicitly supports `LINT_FILES` in its Makefile. Pass `--full` for full Component lint. Focused lint success only validates the selected scope and does not stamp full Component validation; projects without the contract still run full lint. Validation has multiple checks, not modes. `make fix` is a normalize step: complete it first, then run the read-only lint and test checks concurrently against the same stable content. A user may explicitly request only one check; run `run_lint.py` for static quality or `run_tests.py` for behavior, and report the result as partial validation rather than complete validation. The scripts resolve the repo and affected Components, prepare dependencies, run canonical project targets, and update each check's `.devloop` validation stamp. Project targets must follow the [validation contract](references/spec.md). Read only the matching implementation guidance: [Python](references/python.md), [Go](references/go.md), or [Node](references/node.md). Do not add tools, dependencies, or Makefile targets unless requested. Trust the output and report each check's pass/fail. Fix genuine source or test problems; never weaken lint rules, suppress diagnostics, or loosen test assertions merely to force a green result. Only `make fix` may perform automatic source rewrites.