← ClaudeAtlas

speckit-speckit-utils-doctorlisted

Validate project health: templates, agent config, Python runner/helpers, constitution, and feature artifacts
racecraft-lab/racecraft-plugins-public · ★ 5 · Testing & QA · score 74
Install: claude install-skill racecraft-lab/racecraft-plugins-public
# Speckit Utils Doctor Skill ## User Input ```text $ARGUMENTS ``` ## Outline Run a comprehensive health check on the spec-kit project, validating that all required components are present and correctly configured. ### Step 1: Find the project root Locate the `.specify/` directory. If not found, error: "No spec-kit project found. Run `specify init` first." ### Step 2: Check templates Verify these template files exist in `.specify/templates/` and are non-empty: - `spec-template.md` - `plan-template.md` - `tasks-template.md` - `constitution-template.md` - `checklist-template.md` For each template: - File exists and non-empty: PASS - File exists but empty: WARN ("template is empty") - File missing: FAIL ("template not found") ### Step 3: Check agent configuration Read `.specify/init-options.json` to determine the configured AI agent. If no `init-options.json` exists: WARN ("no AI agent configured - run `specify init` to set up") If an agent is configured (e.g., `"ai_assistant": "claude"`): 1. Check that the agent's directory exists (e.g., `.claude/`) 2. Check that command files are registered (e.g., `.claude/commands/speckit.*.md`) 3. Count registered commands - Agent dir exists with commands: PASS - Agent dir missing: FAIL ("agent directory not found") - Agent dir exists but no commands: WARN ("no command files registered") ### Step 4: Check the Python runner and helpers Look for the current SpecKit Pro runner package in either location: - Source checkout, relati