メニュー
Run Vale linting to check for style guide violations. Supports Markdown, AsciiDoc, reStructuredText, HTML, XML, and source code comments. Use this skill when asked to lint, check style, or validate documentation.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | lint-with-vale |
| description | Run Vale linting to check for style guide violations. Supports Markdown, AsciiDoc, reStructuredText, HTML, XML, and source code comments. Use this skill when asked to lint, check style, or validate documentation. |
| model | claude-haiku-4-5@20251001 |
| allowed-tools | Bash, Glob, Read |
Run Vale style linting against documentation files to check for style guide violations.
Before running Vale, check if a .vale.ini or vale.ini exists in the project root:
ls .vale.ini vale.ini 2>/dev/null
If neither exists, create a temporary config in /tmp/ and use it for the lint run:
cat <<'EOF' > /tmp/vale-temp.ini
StylesPath = .vale/styles
MinAlertLevel = suggestion
Packages = RedHat
[*.adoc]
BasedOnStyles = RedHat
[*.md]
BasedOnStyles = RedHat
EOF
Use --config=/tmp/vale-temp.ini when running Vale, and remove the temp config when done.
Inform the user that no project-level .vale.ini was found and a temporary config was used.
Always run vale sync before linting to ensure style packages are up to date:
vale sync
This downloads and updates the style packages defined in .vale.ini (e.g., the RedHat package).
Vale supports many file formats:
.md), AsciiDoc (.adoc, .asciidoc), reStructuredText (.rst), HTML (.html), XML (.xml, .dita).org), plain text (.txt)Run Vale directly against files or directories:
# Single file
vale README.md
vale doc.adoc
vale guide.rst
# Multiple files
vale file1.md file2.adoc file3.rst
# Directory (lints all supported files)
vale docs/
# Specific file patterns
vale --glob='*.md' docs/
vale --glob='*.{md,adoc,rst}' docs/
# Use a specific config file
vale --config=/path/to/.vale.ini docs/
# Match specific file types
vale --glob='**/*.md' path/to/files
vale --glob='**/*.{adoc,md}' path/to/files
# Only show errors and warnings (skip suggestions)
vale --minAlertLevel=warning docs/
# Only show errors
vale --minAlertLevel=error docs/
# JSON output for programmatic use
vale --output=JSON docs/
# Exclude certain files
vale --glob='!**/*-generated.md' docs/
# Lint source code comments
vale --glob='*.py' src/
vale --glob='*.go' pkg/
After running Vale, organize the output for the user:
Lead with errors. Errors are must-fix violations. Present them first with the file, line, and what to change.
Group warnings and suggestions separately. Warnings need attention; suggestions are optional improvements.
Summarize by file when linting multiple files. Show a table with error/warning/suggestion counts per file so the user can prioritize.
Flag likely false positives. Common RedHat Vale false positives include:
When you spot a likely false positive, note it as such rather than recommending the user change correct text.
Do NOT auto-fix source files. Report findings and let the user decide what to change.
docs/guide.md:15:3: error: Style.Spelling - 'kubernetes' should be 'Kubernetes'
docs/guide.md:23:1: warning: Style.PassiveVoice - Avoid passive voice
modules/intro.adoc:45:10: suggestion: Style.SentenceLength - Consider shortening this sentence
Vale must be installed:
# Fedora/RHEL
sudo dnf copr enable mczernek/vale && sudo dnf install vale
# macOS
brew install vale
# Other platforms
# See: https://vale.sh/docs/install
Documentation workflow orchestrator. Reads the step list from .agent_workspace/docs-workflow.yaml (or the plugin default). Runs steps sequentially, manages progress state, handles iteration and confirmation gates. Claude is the orchestrator — the YAML is a step list, not a workflow engine.
Write documentation from a documentation plan. Dispatches the docs-writer agent. Supports AsciiDoc (default) and MkDocs formats. Default placement is UPDATE-IN-PLACE; use --draft for staging area. Also supports fix mode for applying technical review corrections.
Score documentation quality and intent alignment using LLM judge agents (Opus). Dispatches two judge agents in parallel (doc_quality, intent_alignment), extracts specific gaps from AC coverage analysis, and cross-references against scope-req-audit evidence. Produces pass/fail gate with actionable gap list. Iteration logic is owned by the orchestrator, not this skill.
Run code-learner analysis on a source repository for the docs orchestrator workflow. Dispatches a subagent to run learn-code, keeping the heavy orchestration out of the main context. Produces ONBOARDING.md, module registry, per-module summaries, and cross-module relationship data.
Classify JIRA requirements by code evidence status before planning. Uses learn-code analysis data and source code inspection to determine if each requirement is grounded, partial, or absent. Fans out one subagent per requirement for isolated classification. Prevents hallucinated documentation for unimplemented features and surfaces gaps for implemented ones. Conditional on has_source_repo.
Technical accuracy review with optional code-learner claim validation. Iteration logic owned by the orchestrator.