rakib-nyc

Skillassay — AI skill for Claude Code

AI community

Static analyzer for AI coding-agent context — and an Agent Skill your agent can run.

How to install Skillassay

This entry records only its repository, not the path inside it, so there is no exact command to give. Open rakib-nyc/skillassay and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Skillassay does

Static analyzer for AI coding-agent context — and an Agent Skill your agent can run. Measures always-on context cost of CLAUDE.md, AGENTS.md and Agent Skills; finds skills that won't load, redundant instructions, stale paths and duplicate names. Claude Code, Codex, Cursor, Gemini CLI. Research project.

Alternatives in AI

Codeburn — Free, local tool to track AI coding token usage and cost across 37 tools and agents (Claude Code, Cursor, Code 9.7k ★
Axonhub — ⚡️ Open-source AI Gateway — Use any SDK to call 100+ LLMs 3.2k ★
Gentle AI — Agent Skills Index — When working on this project, load the relevant skill(s) BEFORE writing any code 1.9k ★

README

skillassay — measure what your AI coding agent loads before you type

**A static analyzer for AI coding-agent context, and an Agent Skill your agent can run itself.** It measures the always-on context-window cost of `CLAUDE.md`, `AGENTS.md`, `GEMINI.md` and Agent Skills, then reports skills that will not load, redundant instructions, stale path references, duplicate skill names and instruction conflicts — with a citation and a token count on every finding.

Use it two ways:

As a CLI — npx skillassay .
As an Agent Skill — drop it into .agents/skills/ and ask your agent to audit the repo in plain language. It costs 98 always-on tokens.

Works with **Claude Code**, **Codex**, **Cursor** and **Gemini CLI**. Runs offline. No API key.

**This is a research project.** It is published so the method and the numbers can be checked, not as a supported product. Provided **as is, without warranties or conditions of any kind** — see [Disclaimer](#disclaimer).

Install

npx skillassay .

No install, no API key, no config. Requires Node.js ≥ 20.11.

[![npm](https://img.shields.io/npm/v/skillassay)](https://www.npmjs.com/package/skillassay) [![license](https://img.shields.io/npm/l/skillassay)](LICENSE)

Install as an Agent Skill

The package ships an Agent Skill in `.agents/skills/skillassay/` — the cross-client location that **Claude Code, Codex and Gemini CLI all read**, so one file serves every harness:

npm i -D skillassay
mkdir -p .agents/skills
cp -r node_modules/skillassay/.agents/skills/skillassay .agents/skills/

Then just ask, in whichever agent you use:

*"Audit this repo's context cost."* *"Will my new skill actually load?"* *"What's in my always-on context and what can I delete?"*

If you would rather scope it to one client, the same directory works under `.claude/skills/`, `.codex/skills/` or `.gemini/skills/`. `.agents/skills/` is preferred because it is read by all of them.

It costs **98 always-