Fegis

Define AI tools in YAML with natural language schemas. All tool usage is automatically stored in Qdrant vector database, enabling semantic search, filtering, and memory retrieval across sessions.

Open SourceLive

0 0

Add to your client

Add to Cursor Add to VS Code

Claude Code — run this in your terminal:

claude mcp add fegis --env QDRANT_URL=YOUR_QDRANT_URL --env QDRANT_API_KEY=YOUR_QDRANT_API_KEY --env COLLECTION_NAME=YOUR_COLLECTION_NAME --env EMBEDDING_MODEL=YOUR_EMBEDDING_MODEL --env ARCHETYPE_PATH=YOUR_ARCHETYPE_PATH --env AGENT_ID=YOUR_AGENT_ID -- uv --directory /absolute/path/to/fegis run fegis

Claude Desktop and other clients: paste this into the configuration file

{
  "mcpServers": {
    "fegis": {
      "command": "uv",
      "args": [
        "--directory",
        "/absolute/path/to/fegis",
        "run",
        "fegis"
      ],
      "env": {
        "QDRANT_URL": "YOUR_QDRANT_URL",
        "QDRANT_API_KEY": "YOUR_QDRANT_API_KEY",
        "COLLECTION_NAME": "YOUR_COLLECTION_NAME",
        "EMBEDDING_MODEL": "YOUR_EMBEDDING_MODEL",
        "ARCHETYPE_PATH": "YOUR_ARCHETYPE_PATH",
        "AGENT_ID": "YOUR_AGENT_ID"
      }
    }
  }
}

After adding, fill in: QDRANT_URLQDRANT_API_KEYCOLLECTION_NAMEEMBEDDING_MODELARCHETYPE_PATHAGENT_ID

Some arguments are placeholders from the README. Replace them with your own values.

Built from the configuration example in the project's README. The buttons open the app on your computer, which asks you to confirm before anything is added.

Install

git clone https://github.com/p-funk/fegis.git

docker run -d --name qdrant -p 6333:6333 -p 6334:6334 qdrant/qdrant:latest

Copied from the project's README. Read what it does before you run it.

README scanCaution

Its README asks for something worth reading first, such as administrator rights or an install script.

Runs an install script from a well-known hostcurl -LsSf https://astral.sh/uv/install.sh | sh
Needs an API key or token"QDRANT_API_KEY": "",

We scan the README only, not the code. A clean result is not proof of safety. How we scan

Alternatives to Fegis

Listings of the same kind that share the most topics, most-starred first.

Marketing Skills

Claude skills for marketing: 49 free marketing skills for Claude Code and claude.ai (Google Ads, Meta Ads, SEO, AI visibility) + the Ryze Google Ads MCP / Meta Ads MCP: claude mcp add ryze --transport

7.7k

Tradingview Mcp

MCP server for market analysis in AI assistants: screeners, multi-timeframe technical analysis, backtests and sentiment for crypto, stocks, forex and futures. 39 tools. Self-host free, or use the host

5.3k

Unity MCP

AI Skills, MCP Tools, and CLI for Unity Engine. Full AI develop and test loop. Use cli for quick setup. Efficient token usage, advanced tools. Any C# method may be turned into a tool by a single line.

4.6k

Linkedin Mcp Server

Open-source MCP server for LinkedIn. Give Claude and any MCP-compatible AI agent access to profiles, companies, jobs, and messages.

4.2k

Google Workspace Mcp

Control Gmail, Google Calendar, Docs, Sheets, Slides, Chat, Forms, Tasks, Search & Drive with AI - Comprehensive Google Workspace MCP Server & CLI Tool

3.5k

Korean Law Mcp

법령·판례를 AI가 찾고 인용까지 검증하는 MCP 서버 — 법제처 API 42개를 도구 10개로: 법령·조문·별표, 판례·헌재·해석례 23개 도메인, 조례 정비 레이더, 신구조문대비표 | MCP server for Korean law — statutes, precedents, ordinances, comparison tables and citation ve

2.9k

View on GitHub

Repository

Stars: 24
Forks: 7
Last updated: Jul 5, 2025
License: MIT
Language: Python
Created: Apr 4, 2025
Open issues: 0
Quality score: D 33/100

p-funk/fegis

Read from GitHub daily.

See alternatives

Pricing

Open Source

Made by

No owner listed.