workflowstacks
claude-skill

Raven: Boost Your Startup’s AI Ops with a Self‑Improving Agent

Raven lets founders automate deep research and reasoning using a memory‑first, self‑evolving agent built on EverOS. Ideal for teams leveraging Claude, OpenAI, or other LLMs. 5.4k+ GitHub stars.

From the repo · “The Harness of Harnesses • built for RSI: a trusted, persistent, self-evolving multi-agent ecosystem for all-domain collaboration.”

5,543 stars+87★ this week182 forksPythonGuide quality 8/10 · reviewed Oct 14Updated 10/22/2026100% free · open source
What it does

Raven creates self-improving AI agents that utilize EverOS and MiroThinker for deep research and reasoning, enabling founders to build advanced AI models

When to use it
  • •When you need AI agents that can learn and improve over time
  • •When you're working with EverOS and want to integrate MiroThinker's deep research capabilities
  • •When you want to build AI models that can reason and make decisions based on complex data
Ready-to-paste prompt
python raven.py --prompt 'Design a marketing strategy for a new sustainable energy startup'
Heads up: Make sure you have a valid EverOS API key and MiroThinker credentials set up in your config.py file before running the Raven agent
Saves to your device
Use with Claude
New

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

⚡ Runs out of the box

A prompt/skill package — nothing to install beyond adding it to your AI tool.

Try it instantly — no install
Claude Code
mkdir -p ~/.claude/skills/raven && curl -fsSL https://workflowstacks.com/api/skills/raven/claude-skill -o ~/.claude/skills/raven/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 Raven: Boost Your Startup’s AI Ops with a Self‑Improving Agent works
Codeflow
Free to inspect

Raven: Boost Your Startup’s AI Ops with a Self‑Improving Agent is a very large Python project (~648k lines across 2069 code files, plus 1099 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, Apache-2.0 license, has a test suite.

Size
Very large codebase
~648k lines · 2069 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
Python · Node.js · Docker
No API keys detected
Python 77%TypeScript 20%CSS 1%JavaScript 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
    raven/__main__.py
    Where the program starts running
  4. 4
    package.json
    Dependencies and the commands it exposes
  5. 5
    raven/__init__.py
    Inside raven/ — the main logic begins here
What's in each folder
raven/Core code — the actual logic826 files
agents/Prompts, skills & agent definitions151 files
docs/Documentation114 files
demos/Examples you can copy10 files
ui-web/Folder674 files
ui-tui/Folder457 files
plugins-dist/Folder363 files
benchmarks/Evaluations & benchmarks145 files
READMEHas testsDocumentedCI checksDocker readyExamples includedApache-2.0 licenseUpdated this month
Quick Actions
Details
Creator
EverMind-AI
Language
Python
Category
claude-skill
Published
5/21/2026

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/raven.svg)](https://workflowstacks.com/skills/raven?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.

Do this with a working workflow

Pages we wrote and tested that put a skill like this to work — importable n8n templates, outcome guides and Claude Desktop configs.

MCP config

GitHub MCP config

Search repos, read code, open issues and PRs from Claude.

MCP config

Memory (knowledge graph) MCP config

Persistent memory across chats via a local knowledge graph.

MCP config

Sequential Thinking MCP config

Give Claude a structured, step-by-step reasoning scratchpad.

Related skills

More claude-skill tools founders pair with this one.

claude-skill★ 256,115
Hermes-Agent: AI Growth
Scale with a versatile AI agent. For startup founders using Python and AI.
claude-skill★ 187,700
AutoGPT: Build Autonomous AI Agents Faster
AutoGPT lets founders launch self‑directing AI agents without deep ML work, so you can concentrate on product and growth. Ideal for Python‑savvy teams. 188k+ GitHub stars
claude-skill★ 148,235
langchain
The agent engineering platform.
claude-skill★ 4,657
open-knowledge: Turn Markdown into an AI‑Powered Knowledge Hub
A TypeScript‑based editor that lets founders capture notes, build a second‑brain, and query docs with Claude. Perfect for teams that need an LLM‑driven wiki. 4.6k+ GitHub stars.
claude-skill★ 11,079
open-swe
An Open-Source Asynchronous Coding Agent
claude-skill★ 899
Track Claude Code Updates
Founders get prompt tracking and release metadata with claude-code-changelog. For startup founders using Claude Code.