.claude/skills/build-check/SKILL.md
Run build and lint checks, returning only a summary. Build logs are contained within the forked context; only the summary is returned to the main context.
npx skillsauth add peintangos/deep-agents-example build-checkInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Follow these steps to run build and lint checks, outputting only a results summary.
Note: Do not use
run_in_backgroundwith the Bash tool. Run commands directly (foreground) with timeout control. No need to check processes with ps or grep.
Read ralph.toml first if it exists, then CLAUDE.md and docs/architecture.md. Inspect common project files at the repo root only as needed (for example Makefile, justfile, package.json, pyproject.toml, Cargo.toml, go.mod, composer.json, or other task-runner files).
Determine the canonical commands for:
Prefer commands explicitly defined in ralph.toml. If the registry is missing or incomplete, prefer commands explicitly documented in project docs. Only then infer them from the repo's task runner or manifest files.
When ralph.toml exists, use:
build_check for build or compile verificationlint_check for lint or static analysisPrefer repo-level wrappers such as make, just, or documented scripts over framework-specific subcommands when both exist.
If neither check is defined or applicable, report "No build or lint commands defined" and exit.
If a build command exists, run it with a reasonable timeout (for example timeout 300 when available).
If a lint or static-analysis command exists, run it with a reasonable timeout.
Output results in the following format. Do not output the full build log.
## Build & Lint Check Results
- Build: PASS / FAIL / SKIPPED
- Lint: PASS / FAIL / SKIPPED
### Error Details (failures only)
- `filename:line` — error message
- `filename:line` — error message
data-ai
監査レポートを Zenn 技術記事向けの文体 (だ/である調 + 比較表 + 階層化された見出し) に整えるスキル。`src/reporter.ts` の骨組みを LLM が整形してレポート (`out/*.md`) を仕上げるときに読み込む。
testing
OSS リポジトリの既知脆弱性 (OSV / GHSA) を照合し、重大度と影響範囲を分類するスキル。セキュリティ監査を実行するときに読み込む。
data-ai
OSS リポジトリのメンテナンス健全性 (リリース頻度・Issue 対応速度・放置 PR) を定量的に評価するスキル。メンテナンス監査を実行するときに読み込む。
tools
OSS リポジトリのメインライセンスを特定し、商用利用制約・依存互換性・NG ライセンスの検出までを行うスキル。ライセンス監査を実行するときに読み込む。