mcp-server

gini-agent: Create a memory‑rich AI assistant that learns as you work

Name: gini-agent: Create a memory‑rich AI assistant that learns as you work
Author: Open-Curiosity

gini-agent lets founders add a personal, learning AI agent to their apps—built in TypeScript, local‑first, with 2.0k+ GitHub stars.

From the repo · “The agent that remembers and learns.”

2,157 stars+136★ this week900 forksTypeScriptGuide quality 9/10 · reviewed Oct 19Updated 7/18/2026100% free · open source

What it does

The gini-agent is an AI-powered tool that learns and automates tasks for your startup, allowing for more efficient and autonomous operations.

When to use it

•When you need to automate repetitive tasks without manual intervention
•When your startup requires a customized AI agent for specific workflows
•When you want to leverage machine learning for task automation without extensive coding

Ready-to-paste prompt

node src/agent.js --task 'automate_customer_support' --learning_rate 0.1

Heads up: Ensure you have Node.js version 16 or higher installed, as the gini-agent is built with TypeScript and requires this version for compatibility

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/gini-agent && curl -fsSL https://workflowstacks.com/api/skills/gini-agent/claude-skill -o ~/.claude/skills/gini-agent/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How gini-agent: Create a memory‑rich AI assistant that learns as you work works

Codeflow

Free to inspect

Gini-agent: Create a memory‑rich AI assistant that learns as you work is a very large TypeScript project (~290k lines across 799 code files, plus 7 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit 3 months ago, MIT license, has a test suite.

Size

Very large codebase

~290k lines · 799 code files · days to read — use, don't read

Setup

Developer setup

A real software project. Use it via its install path; don't expect to read it all.

Runs on

Node.js · Docker

No API keys detected

TypeScript 99%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
package.json
Dependencies and the commands it exposes
4
skills/README.md
Inside skills/ — an example of what the AI is told to do

What's in each folder

skills/Prompts, skills & agent definitions49 files

packages/web/Frontend / UI260 files

docs/Documentation120 files

packages/runtime/Package465 files

packages/mobile/Package85 files

scripts/Helper scripts5 files

docker/Deployment / infrastructure1 files

patches/Patches1 files

READMEHas testsDocumentedCI checksDocker readyMIT licenseUpdated 3 mo ago

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🛠️ Technical setup

Expect 20–40 minutes in a terminal — or let your AI agent drive it.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/gini-agent && curl -fsSL https://workflowstacks.com/api/skills/gini-agent/claude-skill -o ~/.claude/skills/gini-agent/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

Open-Curiosity

Language

TypeScript