πͺ¨ why use many token when few token do trick β Claude Code skill that cuts 65% of tokens by talking like caveman
Web app and CLI for AI config files
LynxPrompt is a self-hosted platform for creating and managing AI assistant rules, commands, and prompt files. It gives you a central place to generate blueprints, browse shared configs, and export them into the formats your tools expect.

Builders who want one place to manage AI rules and commands across many coding assistants.
You can standardize agent behavior across projects instead of rewriting the same config files by hand.
What it does
Generate configs in a wizard
Create AI config files step by step in the web app or the CLI, with support for repo detection, template variables, and draft saving.
Store and share blueprints
Keep reusable AI configs in a private or federated marketplace so teams can browse, favorite, and reuse them.
Export to many formats
Write once and export to supported assistant formats such as `AGENTS.md`, `CLAUDE.md`, and `.cursor/rules/`.
Use a full CLI
The `lynxp` command can log in, open the wizard, pull blueprints, push local configs, and show CLI settings.
Self-host the stack
Run the app with Docker Compose or Kubernetes, with PostgreSQL, migrations, and environment-based feature flags.
Work with editor and automation integrations
The project includes a VS Code extension, a GitHub Action, and a REST API for scripted workflows.
How to get it
- 1A Helm chart is also available for Kubernetes deployments. See the chart documentationβ¦
helm repo add lynxprompt https://geiserx.github.io/LynxPrompt helm install lynxprompt lynxprompt/lynxprompt
README

LynxPrompt
Self-hostable AI config management for teams and individuals
What is LynxPrompt?
LynxPrompt is a self-hostable platform for managing AI IDE configuration files β AGENTS.md, .cursor/rules/, CLAUDE.md, slash commands, and 30+ other formats. Deploy it on your own infrastructure and give your team a central hub to create, share, and standardize AI coding assistant configurations across every project.
Instead of manually writing configuration files for every project and every AI tool, use LynxPrompt to:
- Generate configs through an interactive wizard (web or CLI)
- Share blueprints through a private or federated marketplace
- Standardize AI behavior across projects with reusable templates
- Export to any supported format with one click
LynxPrompt is free and open-source. Self-host it for personal use, or deploy it within your organization to enforce coding standards, share institutional knowledge, and ensure consistent AI assistant behavior across your engineering teams. A hosted instance is also available at lynxprompt.com for those who prefer not to self-host.
Key Features
Universal AI Config Hub
Supports 30+ AI coding assistants β Cursor, Claude Code, GitHub Copilot, Windsurf, Zed, Aider, Gemini CLI, Cline, Roo Code, Amazon Q, JetBrains Junie, and many more. Write once, export to any format.
Blueprint Marketplace
Internal or federated marketplace for sharing AI configurations and slash commands within your organization. Browse, search, favorite, and reuse blueprints across teams.
Interactive Wizard
Step-by-step config generator available on both web and CLI. Auto-detects your tech stack, frameworks, and repo structure from GitHub/GitLab URLs. Supports template variables, monorepo hierarchies, and draft auto-saving.
Configurable Authentication
Flexible auth to fit your environment:
- OAuth β GitHub, Google
- Email β Passwordless magic link login
- Passkeys β WebAuthn biometric/hardware key authentication
- SSO β SAML, OIDC, and LDAP for enterprise identity providers
Optional AI-Powered Editing
Enable AI-assisted blueprint creation and editing with your own Anthropic API key. Entirely optional β works fully without it.
Full REST API + CLI Tool
Programmatic access for automation and CI/CD integration. Generate API tokens, fetch blueprints, search, and download via REST. The CLI (lynxp) mirrors the full web platform feature set.
Self-Hostable with Docker Compose
Single docker compose up to run the entire stack. PostgreSQL included. Auto-runs database migrations on startup. Toggle every feature via environment variables.
Quick Start (Self-Hosting)
# 1. Create a .env file
cat > .env <<EOF
NEXTAUTH_SECRET=$(openssl rand -base64 32)
ADMIN_EMAIL=your@email.com
APP_URL=http://localhost:3000
EOF
# 2. Download the self-host compose file and start LynxPrompt
curl -O https://raw.githubusercontent.com/GeiserX/LynxPrompt/main/docker-compose.selfhost.yml
docker compose -f docker-compose.selfhost.yml up -d
# 3. Open http://localhost:3000
That's it. LynxPrompt is running with PostgreSQL, automatic migrations, and email authentication enabled by default.
Helm Chart (Kubernetes)
A Helm chart is also available for Kubernetes deployments. See the chart documentation for the full values reference.
helm repo add lynxprompt https://geiserx.github.io/LynxPrompt
helm install lynxprompt lynxprompt/lynxprompt
Configuration
All features are controlled via environment variables. Toggle what you need, disable what you don't.
| Variable | Default | Description |
|---|---|---|
NEXTAUTH_SECRET | (required) | Session encryption key |
APP_URL | http://localhost:3000 | Base URL of your instance |
APP_NAME | LynxPrompt | Instance name shown in the UI |
ENABLE_GITHUB_OAUTH | false | GitHub OAuth login |
ENABLE_GOOGLE_OAUTH | false | Google OAuth login |
ENABLE_EMAIL_AUTH | true | Magic link email login |
ENABLE_PASSKEYS | true | WebAuthn passkey authentication |
ENABLE_TURNSTILE | false | Cloudflare Turnstile CAPTCHA |
ENABLE_SSO | false | SAML / OIDC / LDAP authentication |
ENABLE_USER_REGISTRATION | true | Allow public sign-ups |
ENABLE_AI | false | AI-powered editing features |
AI_MODEL | claude-3-5-haiku-latest | AI model for editing |
ANTHROPIC_API_KEY | Required when ENABLE_AI=true | |
ENABLE_BLOG | false | Blog module |
ENABLE_SUPPORT_FORUM | false | Support forum module |
ENABLE_STRIPE | false | Stripe payments for marketplace |
SUPERADMIN_EMAIL | Auto-promote this email to superadmin | |
APP_LOGO_URL | Custom logo URL | |
UMAMI_SCRIPT_URL | Umami analytics script URL | |
CONTACT_EMAIL | Contact form destination | |
STATUS_PAGE_URL | Status page link |
CLI
The CLI tool mirrors the web platform and works against any LynxPrompt instance. By default it connects to lynxprompt.com, but you can point it to any self-hosted deployment.
# Install
npm install -g lynxprompt
# (Optional) Point to a self-hosted instance β two ways:
lynxp config set-url https://lynxprompt.your-company.com
# or: export LYNXPROMPT_API_URL=https://lynxprompt.your-company.com
# Authenticate (opens browser on the configured instance)
lynxp login
# Generate AI config files interactively
lynxp wizard
# Pull a blueprint
lynxp pull bp_abc123
# Push local configs
lynxp push
# View current CLI configuration
lynxp config
The API URL is stored in the CLI config file (see lynxp config path). The LYNXPROMPT_API_URL environment variable takes precedence if set.
Also available via Homebrew (brew install GeiserX/lynxprompt/lynxprompt) and Chocolatey (choco install lynxprompt).
VS Code Extension
Prefer managing configs without leaving the editor? LynxPrompt also has an official VS Code extension.
From inside VS Code you can:
- Browse your cloud blueprints and local AI config files in a dedicated sidebar
- Pull
AGENTS.md,CLAUDE.md,.cursor/rules/, and more into the correct workspace paths - Diff local files against their cloud versions with the built-in editor
- Push updates back to LynxPrompt without leaving VS Code
Install it from the Marketplace or run ext install LynxPrompt.lynxprompt.
Architecture
- Frontend + API: Next.js 16 with App Router
- Database: PostgreSQL with Prisma ORM
- Deployment: Docker Compose with auto-migration on startup
- Auth: NextAuth.js with configurable providers
- Search: Full-text search via PostgreSQL
Supports single-database or multi-database setups depending on your scale requirements.
Development
git clone https://github.com/GeiserX/LynxPrompt.git
cd LynxPrompt
cp env.example .env
docker compose up -d
npm install --legacy-peer-deps
npm run dev
See CONTRIBUTING.md for contribution guidelines.
Documentation
Full documentation is available at lynxprompt.com/docs, covering:
- Getting Started
- Configuration Wizard
- Blueprints & Commands
- CLI Reference
- API Reference
- Self-Hosting Guide
Ecosystem
| Project | Type | Description |
|---|---|---|
| lynxprompt-mcp | MCP Server | Browse and manage blueprints from AI assistants via the Model Context Protocol |
| lynxprompt-vscode | VS Code Extension | Pull, diff, and push AI config files without leaving the editor |
| homebrew-lynxprompt | Homebrew Tap | Install the LynxPrompt CLI on macOS/Linux via brew install GeiserX/lynxprompt/lynxprompt |
| n8n-nodes-lynxprompt | n8n Community Node | Automate blueprint workflows in n8n |
License
This project is licensed under the GNU General Public License v3.0.
Author: Sergio FernΓ‘ndez Rubio (GeiserX)
Files in the repo
- .github
- action
- charts
- cli
- docs
- packages
- prisma
- public
- scripts
- snap
- src
- test
- tests
- .coderabbit.yaml
- .dockerignore
- .ghost-portfolio.yml
- .gitattributes
- .gitignore
- .pre-commit-config.yaml
- .prettierrc
- CHANGELOG.md
- CLAUDE.md
- codecov.yml
- CONTRIBUTING.md
- docker-compose.selfhost.yml
- docker-compose.yml
- Dockerfile
- entrypoint.sh
- env.example
- eslint.config.mjs
- LICENSE
- next.config.ts
- package-lock.json
- package.json
- postcss.config.mjs
- README.md
- RELEASING.md
- SECURITY.md
- tsconfig.json
- vitest.config.ts
Discussion (0)
Ask about usage, or say what you built with itSign in to join the discussion.
No comments yet. Be the first to say what this is good for.
More tools
The best-benchmarked open-source AI memory system. And it's free.
Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and remote runtime.

A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io
Never stop coding. Free MIT AI gateway: one endpoint, 352 providers (150+ free), 1200+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 550+ contributors
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.