rely-ai-org

rely-ai-org / caliber

Public

Improve your agentic development experience with one command. Generate optimized AI agent configs for Claude Code, Cursor, and Codex.

www.npmjs.compackage@rely-aicaliber agent-config ai-agents anthropic claude-code claude-md
15
2
100% credibility
Found Mar 16, 2026 at 15 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Caliber scans your codebase to create customized setup files for AI coding tools like Claude Code, Cursor, and OpenAI Codex, keeping them in sync as your project evolves.

How It Works

1
🔍 Discover Caliber

You find a helpful tool that makes your AI coding helpers understand your project perfectly.

2
📦 Get it set up

With one simple download, Caliber is ready to explore your code.

3
Pick your AI buddy
🧠
Claude Code

Connect your Claude subscription for powerful code insights.

✨
Cursor

Link your Cursor setup for seamless editing help.

⚡
Other AI

Use your own AI service for custom thinking power.

4
🔬 Caliber scans your project

It studies your files, patterns, and setup to learn everything about your work.

5
✏️ Review the magic

See smart suggestions tailored just for you and make any tweaks.

🎉 AI supercharged!

Your coding assistant now knows your project like the back of its hand, saving you time every day.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 15 to 15 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is caliber?

Caliber is a TypeScript CLI app that scans your codebase—languages, frameworks, deps, structure—and spits out tailored configs for Claude Code, Cursor, and Codex with one command: `caliber init`. It generates CLAUDE.md docs, skills in OpenSkills format, Cursor rules, and MCP servers, fixing stale refs or drift to aide agentically improve visual language model performance with domain experts. No API key needed if you have Claude Code or Cursor seats; just npm install -g and go.

Why is it gaining traction?

Unlike manual config drudgery, it scores your setup A-F deterministically (no LLM calls), auto-refreshes via git pre-commit or session-end hooks, and lets you chat-refine outputs. Cross-tool parity keeps Claude, Cursor, Codex consistent, boosting agentic AI how-to-improve tips while slashing onboarding time—perfect for github improve code quality without rewriting prompts endlessly. Reversible backups and dry runs make it zero-risk.

Who should use this?

AI-heavy teams on Claude Code or Cursor building TypeScript/Python/Go apps, especially those tired of generic AI outputs ignoring project conventions. Solo devs or new hires onboarding to monorepos will love the instant context; use it to improve github copilot performance analogs and github improve readme for agent skills.

Verdict

Grab it if you're deep in agentic workflows—early promise with solid docs and tests, despite 15 stars and 1.0% credibility signaling newness. Run `caliber score` first to gauge fit; MIT license means fork-friendly upside.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.