sheil-ai

[](https://sdk.vercel.ai) [](https://anthropic.com) [](https://gemini.google.com) [](https://openai.com) [](https://ollama.com) [](https://openrouter.ai) [](https://www.postgresql.org) [](https://github.com/pgvector/pgvector) [](https://opensource.org/licenses/MIT)

a modular ai cli that unifies multiple model providers with built-in rag, embeddings, pluggable tools, and mcp support.

• multi-provider support: anthropic (claude), google (gemini), openai (gpt), ollama, openrouter
• rag system: vector-based knowledge retrieval with postgresql + pgvector
• mcp support: context7, exa, and filesystem integration, extendable via config
• comprehensive toolset: web search, shell execution, calculations, http requests, uuid generation
• interactive chat: session management with conversation history
• knowledge base: web scraping and content ingestion pipeline
• configurable parameters: temperature, tokens, reasoning modes, tool usage

• node.js v24.7.0+ (see .nvmrc)
• postgresql with pgvector and uuid-ossp extension
• docker and docker compose (for migrations)

# clone repository
git clone https://github.com/mcking-07/sheil-ai.git
cd sheil-ai

# install dependencies
npm install

# set up environment
cp .env.example .env
# edit .env with your api keys

# start postgresql w