Happy slice of mcp pizzaMCP.pizza

skynet-agent

MCP.Pizza Chef: esinecan

This is a chat app you run yourself in a browser, with the model of your choice behind it: Google, OpenAI, Anthropic, DeepSeek, Groq, Mistral, or a local one through Ollama. Old conversations stay searchable, anything important can be saved with tags and a priority score, and links between saved items are tracked so you can ask what relates to what. It also connects out to other MCP servers for extra tools, and an autopilot mode chases a goal over several turns.

Data
Notes

Use This MCP client To

Chat with a model that remembers last week's discussion Save an important detail so it resurfaces later Search my past conversations for something I half remember Switch between different models without changing apps Let it work through a long goal over several turns Keep my chats and saved notes on my own machine

README

Skynet Agent

What if AI could not only access memories, but consciously choose what to remember? With MCP tool access fully supported? image

TypeScript Next.js ChromaDB MCP

AI conversation platform implementing dual-layer memory architecture inspired by human cognition. Combines automatic background memory with conscious, deliberate memory operations that AI controls. Tool access powers similar to Claude Desktop.

Core Features

LangGraph-Powered Autopilot

Purpose-driven autonomous execution replacing simple query generation with sophisticated multi-step workflows:

  • Purpose analysis and strategic planning
  • Context gathering from all memory systems
  • Smart tool orchestration with error recovery
  • Progress monitoring with adaptive replanning
  • Reflection engine for continuous learning
  • Configurable aggressiveness and safety controls

skynet-agent FAQ

Is this something I add to Claude, or its own app?
It is its own chat app that you run on your own computer, and it connects out to MCP servers for extra tools.
Do I need an account key?
Yes, one for whichever model provider you pick. The free Google AI Studio option is the suggested starting point, or run a local model through Ollama.
How hard is setup?
This one is for comfortable tinkerers. You copy the project, start two databases with Docker, and run it from a terminal.
Can I use this to keep notes my assistant looks up later?
Yes. Tell it to remember something, tag it, mark how important it is, and search for it in a later chat.
Which models can I use?
Google, OpenAI, Anthropic, DeepSeek, Groq, Mistral, and local models through Ollama.
Where does my data live?
The memory stores run on your own machine through Docker, though your messages still travel to whichever model provider you chose.