xiaolaiCommunity·10 Agent Skills Included

nlpm-for-claude

Score, lint, fix, and test natural-language instruction files

Scores skills, agents, commands, rules, hooks, and prompts on a 100-point quality rubric across Claude Code, Codex CLI, and Antigravity. Catches vague wording, broken cross-references, missing manifest entries, and security risks that manual review misses. Auto-fixes mechanical issues, tracks quality trends, and runs test specs so instruction files stay reliable as projects grow.

111|29

↗

npx skills add xiaolai/nlpm-for-claude --all -g -y

Available:

Tells the agent how each slash command dispatches scanner, scorer, checker, and tester sub-agents, and which reference skills each agent loads when scoring or fixing instruction files.

All Skills in This Repository (10)

Pure Emerald Level Indicators

📦 In Repo

xiaolai

writing-hooks

Design and validate Claude Code hooks with event types and matcher patterns.

writing-rules

Automate creation and governance of CLAUDE rules in .claude/rules.

writing-plugins

Document architectural best practices for Claude Code plugin development and publishing.

writing-agents

Define authoring standards for Claude Code agents with structured output schemas.

testing

Automate test-driven development workflows for NL artifacts using .nlpm-test spec files.

writing-prompts

Structure prompts with a five-layer Role, Context, Task, Constraints, and Output framework.

orchestration

Coordinates multi-agent workflows for Claude Code deployments.

scoring

Compute 100-point quality scores for NL programming artifacts using predetermined penalties.

rules

Enforce R01–R50 quality rules on NL programming artifacts.

writing-skills

Standardize SKILL.md files with frontmatter schema and structured body sections.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install nlpm-for-claude?▼

Run `npx skills add xiaolai/nlpm-for-claude --all -g -y` in your terminal to install all skills in this collection globally.

What does nlpm actually check?▼

It scores skills, agents, commands, rules, hooks, and prompts against 50 quality rules, verifies cross-file consistency, and flags security risks in executable files like hooks and scripts.

Does nlpm work with Codex CLI and Antigravity?▼

Yes. The scoring rubric covers Claude Code, Codex CLI, and Antigravity through per-tool convention overlays, and a standalone Python validator runs in CI without any of these tools installed.

Can nlpm fix problems automatically?▼

Yes. The fix command repairs mechanical findings automatically, while judgment-based issues are reported with concrete suggestions for you to apply.

Do I need coding experience to use nlpm?▼

No. You run simple commands like score or check, and the agent reads your instruction files and reports a clear quality score with specific fixes.