Repolode
Submit
Gentleman-Programming

Gentleman-Programming/gentle-ai

Configures AI coding agents like Claude Code, Cursor, OpenCode and Codex with persistent memory, shared workflows, skills and MCP servers.

View on GitHubWebsite
Stars
8,496
Forks
884
Open issues
972
Last commit
20 hours ago

What is it?

Gentle-AI is a Go CLI that sets up the AI coding agents you already use, including Pi, OpenCode, Claude Code, Codex, Cursor and VS Code Copilot. It installs a shared configuration covering persistent memory through Engram, a skills library that loads when a task matches, optional MCP servers such as Context7, and selectable personas and themes. It does not install the agents themselves.

The project's main workflow, Organic Driven Development, sizes work by understanding and risk rather than file count: small changes run inline, while substantial tasks keep one recoverable feature document. A separate review mode, Receipt-Driven Development, freezes the exact candidate change before a read-only review, so the findings match the version you plan to commit. A `gentle-ai doctor` command reports health without changing files, and configs are backed up before each write.

Gentle-AI is aimed at developers who use several coding agents and want consistent memory, workflow and review habits across them. It is MIT licensed and runs on macOS, Linux and Windows.

Key features

  • 01Configures 17 AI coding agents, including Pi, OpenCode, Claude Code, Codex and Cursor.
  • 02Persistent project memory via Engram, so decisions carry across sessions.
  • 03Organic Driven Development keeps small tasks inline and large tasks resumable.
  • 04Receipt-Driven Development reviews a frozen snapshot of the change.
  • 05Security deny-list blocks access to SSH keys, .env and credential files.
  • 06Config backups are taken before every write, and `doctor` checks health read-only.

Use cases

  • Keeping one set of memory, skills and workflow rules across several coding agents.
  • Giving an agent project context so it stops re-asking the same questions each session.
  • Adding a structured review step before committing AI-written changes.
  • Standardizing how a team uses AI coding assistants.

Quick start

  1. 1

    Follow the quickstart guide linked in the README (docs/quickstart.md)

In collections

Claude Code tools, skills & pluginsOpen-source AI agent frameworks & appsAI coding assistants & agentsMCP serversAI agent skill libraries

Topics

#ai-agents#ai-coding#coding-agents#persistent-memory#mcp#agent-skills#code-review#developer-tools#claude-code#codex

FAQ

What is gentle-ai?
Configures AI coding agents like Claude Code, Cursor, OpenCode and Codex with persistent memory, shared workflows, skills and MCP servers. Gentle-AI is a Go CLI that sets up the AI coding agents you already use, including Pi, OpenCode, Claude Code, Codex, Cursor and VS Code Copilot. It installs a shared configuration covering persistent memory through Engram, a skills library that loads when a task matches, optional MCP servers such as Context7, and selectable personas and themes. It does not install the agents themselves.
Is gentle-ai free to use? What license does it have?
gentle-ai is open source under the MIT license. Check the license terms before using it in commercial products.
Is gentle-ai still maintained?
Actively maintained. The last commit to gentle-ai was on Nov 6, 2026.
What can I use gentle-ai for?
Keeping one set of memory, skills and workflow rules across several coding agents. Giving an agent project context so it stops re-asking the same questions each session. Adding a structured review step before committing AI-written changes. Standardizing how a team uses AI coding assistants.
How do I install gentle-ai?
Follow the quickstart guide linked in the README (docs/quickstart.md)
Where can I get gentle-ai?
The source code and installation instructions are on GitHub: https://github.com/Gentleman-Programming/gentle-ai

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects