mcp-server

chat-on-steroids

Cross-platform local MCP capabilities for ChatGPT with Chrome integration, Goal, Compact & Resume, and durable multi-agent workflows.
4,257 stars+25★ this week568 forksTypeScriptUpdated 10/9/2026100% free · open source
What it does

Runs a local multi‑agent server that lets Chrome‑based extensions talk to ChatGPT with goal‑driven, resumable, and compact workflows.

When to use it
  • •You need a self‑hosted ChatGPT back‑end that can keep state across browser sessions.
  • •Your product requires multiple AI agents to coordinate (e.g., summarizer + planner) without hitting external rate limits.
  • •You want Chrome extensions to trigger AI actions locally, preserving privacy and reducing latency.
Ready-to-paste prompt
In the Chrome extension console, run: `chrome.runtime.sendMessage({ action: 'runGoal', goal: 'Create a 3‑step launch plan for a new SaaS product', resumeId: null })` – the server will orchestrate the multi‑agent workflow and return a compact, resumable result.
Heads up: The server will fail to start if `OPENAI_API_KEY` is missing or invalid; the first request will return a 401 error.
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/ba138273-023f-4523-9c7b-a07c893364fb && curl -fsSL https://workflowstacks.com/api/skills/ba138273-023f-4523-9c7b-a07c893364fb/claude-skill -o ~/.claude/skills/ba138273-023f-4523-9c7b-a07c893364fb/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 chat-on-steroids works
Codeflow
Free to inspect

Chat-on-steroids is a very large TypeScript project (~123k lines across 289 code files, plus 241 test files). It is a full software project: use it through its install path rather than reading it end to end. Last commit this month, MIT license, has a test suite.

Size
Very large codebase
~123k lines · 289 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
No API keys detected
TypeScript 82%JavaScript 15%CSS 2%HTML 1%Swift 1%
Where to start reading
  1. 1
    README.md
    Start here — what it does and how to install it
  2. 2
    AGENTS.md
    The instructions the AI actually follows
  3. 3
    package.json
    Dependencies and the commands it exposes
  4. 4
    src/main/index.ts
    Inside src/ — the main logic begins here
What's in each folder
src/Core code — the actual logic249 files
extension/Sub-packages / plugins17 files
docs/Documentation272 files
scripts/Helper scripts68 files
native/Folder3 files
artwork/Folder2 files
test/Tests — proof it works239 files
.github/CI / automation (GitHub Actions)7 files
READMEHas testsDocumentedCI checksMIT licenseUpdated this month
Quick Actions
Details
Creator
totec448-spec
Language
TypeScript
Category
mcp-server
Published
8/22/2026

Are you the creator of this tool? and earn 85% of every sale.

Show it off in your README:

Featured on WorkflowStacks
[![Featured on WorkflowStacks](https://workflowstacks.com/api/badge/ba138273-023f-4523-9c7b-a07c893364fb.svg)](https://workflowstacks.com/skills/ba138273-023f-4523-9c7b-a07c893364fb?utm_source=github&utm_medium=badge)
🔥 Hot this week, in your inbox
Skills like this, every Monday.

The five fastest-growing open-source AI skills, ranked by GitHub star growth. One email a week, unsubscribe anytime.