Ajolote AI — AI skill for Claude Code
Shared AI agent configuration for multi-tool teams.
How to install Ajolote AI
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open antoumpas/ajolote-ai and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Ajolote AI does
Shared AI agent configuration for multi-tool teams. ajolote keeps a single canonical config (.agents/config.json) and translates it into the native format of each developer's AI coding assistant — Claude Code, Cursor, Windsurf, GitHub Copilot, Cline, and Aider. Tool-specific files are gitignored; only the shared config is committed.
Alternatives in AI
- Part 4 — Generate AGENTS.md And AI Agent Configuration Files — I'll help you create the instruction files that will guide your AI coding assistant to build your MVP 2.1k ★
- Paca — AI-native, free, open-source alternative to Jira, Trello, ClickUp & Monday 1.8k ★
- Wax — Shared Single-file memory layer for all your agents, sub mili-second RAG over text, photo and video on Apple S 791 ★
README
ajolote
One shared AI agent config for teams where developers use different tools.
You commit `.agents/` — a single directory of rules, skills, personas, and MCP server definitions. Every developer runs `ajolote use ` and gets a freshly generated config in their tool's native format. No more per-tool config files committed to git, no more "it works on my Cursor but not your Claude Code."
Works with any language or framework. Supports Claude Code, Cursor, Windsurf, GitHub Copilot, Cline / Roo Code, Aider, Gemini CLI, and Codex CLI. Also generates a committed `AGENTS.md` (Linux Foundation / AAIF standard) that any compatible tool can read without ajolote installed.
**When it helps most:**
- Your team uses different tools (Alice is on Claude Code, Bob is on Cursor, CI runs Copilot)
- You want a single place to evolve agent rules without touching per-tool config files
- You need MCP servers shared across tools, with personal tokens staying local and out of git
- You want
ajolote diffin CI to catch config drift before it reachesmain
The mental model
| Committed to git ✅ | Gitignored — generated locally ❌ |
|---|---|
.agents/config.json — the canonical config |
CLAUDE.md, .claude/settings.json |
.agents/rules/*.md — your ground rules |
.cursor/rules/agents.mdc, .cursor/mcp.json |
.agents/skills/*.md — reusable instructions |
.windsurf/rules/agents.md |
.agents/personas/*.md — role behaviours |
.github/copilot-instructions.md |
.agents/context/*.md — project knowledge |
.clinerules, .roo/mcp.json, .roomodes |
.agents/commands/*.md — shared slash commands |
.aider.conf.yml, GEMINI.md |
AGENTS.md — AAIF standard (any tool reads it) |
.codex/config.toml |
Everyone edits the `.agents/` files. Each developer runs `ajolote use ` once to generate their own tool config from the shared source of truth.
Install
**macOS / Linux**
curl -fsSL https://raw.githubusercontent.com/antoumpas/aj
Related Skills
AI Sync
o CLI that keeps a canonical .ai source of rules, skills, agents, prompts, and MCP config, then generates sync
AI Coding Agents
Single source of truth for AI coding agent configuration — skills, commands and rules shared across OpenCode a
Copilot Setup
Universal AI coding assistant configuration engine. Discovers, merges, and deploys config from multiple source
Claude Code Global Config
Production-ready global CLAUDE.md template for Claude Code CLI - Anthropic's AI coding assistant. Clean Archit
Aitoolplus
A high-performance native AI coding assistant configuration workbench built with Rust and GPUI. Supports Claud
Skillr
Write AI instructions once, sync to every tool. One canonical format (.skillr/) compiles to CLAUDE.md, .cursor
Related Agents
Config Workbook Author
Compile the canonical 10-section Salesforce Configuration Workbook. SfSkills admin workflow agent, /author-con
Development Orchestrator Agent
Owns everything between contract lock and Gate 2 — branch verification, conditional shared-package-agent, para
LinkedIn Post Writer
Drafts LinkedIn posts from a single idea. Reads the idea's content from the user or from Notion, selects forma