xtarterize
Apply production-grade conformance configurations (linting, type checking, CI, editor settings, AI skills, release tooling) to JS/TS projects. Scans package.json and config files to auto-detect framework, bundler, and styling, then resolves applicable tasks. Use when user mentions "xtarterize", "conformance", "setup linting", "add CI", "configure project", "init xtarterize", or any task name like "lint/biome", "ts/strict", or "ci/workflow".
- License 许可
- MIT
- 最近更新
- 2026-11-07
- Stars
- ★ 3
- Forks
- 1
01 前言 Front Matter
- name
- xtarterize
- description
- Apply production-grade conformance configurations (linting, type checking, CI, editor settings, AI skills, release tooling) to JS/TS projects. Scans package.json and config files to auto-detect framework, bundler, and styling, then resolves applicable tasks. Use when user mentions "xtarterize", "conformance", "setup linting", "add CI", "configure project", "init xtarterize", or any task name like "lint/biome", "ts/strict", or "ci/workflow".
02 资源文件 2
- 参考文档references/commands.md11.0 KB
- 参考文档references/tasks.md9.3 KB
03 SKILL.md 正文
xtarterize
Scans any JS/TS project and applies curated configs via a task-based engine. Always use --json for agent-driven invocations.
When this skill loads
This skill is activated when the user asks about project conformance, linting setup, CI, or any xtarterize task. Upon loading:
- Assess the ask - is the user requesting a full setup (
init), a single task (add), a status check (check), a task search (query), or diagnostics (doctor)? - Check current state first - run
check --jsonto understand what's already configured before proposing changes - Never guess task IDs - load the task reference before using
add - Always use
--json- parse structured output, never terminal text
Quick reference
npx xtarterize check --json # { ok, summary, tasks: [{ id, label, group, status }], diagnostics }
npx xtarterize diff --json # { files: [{ filepath, stats }], ok, summary }
npx xtarterize list --json # { profile, tasks }
npx xtarterize query "strict typescript" --json # { type: "query", results: [...] }
npx xtarterize doctor --json # { diagnostics: [{ name, status, message }] }
npx xtarterize add lint/biome --json
npx xtarterize init --json --yes
JSON parsing patterns
After each command, parse the JSON to decide the next action:
| Command | Parse this field | To decide |
|---|---|---|
check --json |
tasks[].status |
If any status is "new" or "patch", run init or add |
check --json |
diagnostics[].status |
If any is "fail", run doctor before proceeding |
diff --json |
files, ok |
Empty files (or ok: true) means nothing to change; otherwise show files[].filepath and files[].stats.added/stats.removed as the change magnitude |
list --json |
tasks[].status, profile |
Find the right task ID to pass to add; understand the detected stack (framework, bundler, etc.) |
doctor --json |
diagnostics[].status |
Any "fail" needs fixing; "warn" is advisory |
init --json |
ok |
If false, something went wrong - check stderr |
query --json |
results[].relevance |
Score >= threshold means relevant. Use results[].signals to see which fields matched strongest |
query --json |
results[].taskId |
Pass to add to apply the matched task |
query --json |
count |
If 0, no tasks met the threshold - broaden query or lower --threshold |
Task statuses are new, patch, skip, and conflict; the parsing table above shows how to act on each.
Agent workflows
Initialize a project
# 1. Check state
out=$(npx xtarterize check --json --cwd "$dir")
# Parse out.tasks[].status - if all "skip", skip init
# 2. Preview
out=$(npx xtarterize diff --json --cwd "$dir")
# Parse out - if empty array, nothing to do
# 3. Apply
out=$(npx xtarterize init --json --yes --cwd "$dir")
Add one task
# MANDATORY - Load references/tasks.md first to find exact task ID
out=$(npx xtarterize add ts/strict --json --cwd "$dir")
Diagnose failures
out=$(npx xtarterize doctor --json --cwd "$dir")
# Parse out.diagnostics - any { "status": "fail" } needs attention
# Common fails: "Biome not installed", "Node version too old"
Error handling
| Signal | Likely cause | Fix |
|---|---|---|
add <id> → "Task not found" |
Wrong task ID | list --json, parse tasks[].id |
Unknown option --jsn ... Did you mean --json? |
Typoed flag or command | Follow the suggestion, or run the command with --help for valid options |
doctor shows tool not installed |
Missing dep | xtarterize only writes configs; user may need to install separately |
Task shows "conflict" |
Config differs | Never auto-apply - present to user, only with --include-conflicts |
--cwd fails preflight |
No package.json |
Verify path exists and is a JS/TS project |
query returns empty results |
Query too narrow or no matching tasks | Broaden the query or lower --threshold |
Anti-patterns
- NEVER run
init --yeswithoutdiff --jsonfirst on existing projects - NEVER edit
.xtarterize/backups/- userestoreto recover - NEVER use
--include-conflictswithout user approval - NEVER parse terminal output - always use
--json - NEVER guess task IDs - load references/tasks.md first
Reference files
- Load references/tasks.md before
add- find exact task IDs. MANDATORY - READ ENTIRE FILE. - Load references/commands.md for detailed flag descriptions and JSON output shapes. Do NOT load for basic usage - covered above.
04 同仓库更多 Skills
create-xtarter-app
Scaffold new JS/TS projects from curated starter templates pre-configured with Biome, TypeScript strict mode, Git hooks, GitHub Actions workflows, VS Code settings, and AI agent skills. Use when user wants to create a new project, scaffold an app from a template, or mentions "create-xtarter-app", "npx create-xtarter-app", "scaffold", "starter template", or a template name like "next-tailwind", "vite-chakra", "vite-tailwind", or "next-chakra".
write-a-skill
Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
template-skill
Replace with description of the skill and when Claude should use it.
skill-creator
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
opensrc
Fetch dependency source code to give AI agents deeper implementation context. Use when the agent needs to understand how a library works internally, read source code for a package, fetch implementation details for a dependency, or explore how an npm/PyPI/crates.io package is built. Triggers include "fetch source for", "read the source of", "how does X work internally", "get the implementation of", "opensrc path", or any task requiring access to dependency source code beyond types and docs.
improve-codebase-architecture
Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.