Foundry
The foundation layer for agentic intelligence.
What is Foundry?
Foundry is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to foundation layer for agentic intelligence.
The foundation layer for agentic intelligence.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- The foundation layer for agentic intelligence.
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx foundryConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Foundry
Foundry (promptise) is a Python agent framework that provides the foundation layer for building autonomous AI agents with built-in safety, memory, and reasoning capabilities. It exposes 14 meta-tools that let agents discover and invoke MCP servers, manage conversation memory with vector storage, enforce security policies (PII redaction, injection blocking), apply semantic caching, and run structured reasoning patterns such as ReAct, debate, and research pipelines. Developers install it as a Python library and configure agents programmatically, making it well-suited for teams building production-grade multi-agent systems.
Prerequisites
- Python 3.10 or later installed
- pip or a virtual environment manager (venv, uv, or poetry)
- An API key for at least one supported LLM provider (e.g. OpenAI, Anthropic)
- An MCP-compatible client or the ability to run the agent server directly from Python
Install the promptise package
Install Foundry from PyPI using pip. It is published as the 'promptise' package.
pip install promptiseConfigure an LLM provider
Specify your model in the Foundry string format. The package supports OpenAI, Anthropic, and other providers using the 'provider:model-name' syntax.
# Example: set your API key in the environment
export OPENAI_API_KEY="sk-..."
# or
export ANTHROPIC_API_KEY="sk-ant-..."Set up a memory provider (optional)
Foundry supports ChromaDB for persistent vector memory. Initialize a ChromaProvider with a local persist directory to give your agent long-term memory.
from promptise.memory import ChromaProvider
memory = ChromaProvider(persist_directory="./agent_memory")Build and run an agent
Create an agent with automatic MCP tool discovery, memory injection, and safety scanning enabled. The agent resolves all MCP tools at startup and exposes them for use.
from promptise import Agent
agent = Agent(
model="openai:gpt-4o-mini",
memory=memory,
mcp_servers=["npx @modelcontextprotocol/server-filesystem /tmp"],
safety=True
)
response = await agent.run("List the files in /tmp and summarize their sizes.")Connect as an MCP server
Expose the Foundry agent as an MCP server so Claude Desktop or other clients can invoke it as a tool. Use the HTTPServerSpec to configure the transport.
from promptise import Agent
from promptise.transport import HTTPServerSpec
agent = Agent(model="openai:gpt-4o-mini")
agent.serve(HTTPServerSpec(host="127.0.0.1", port=8765))Foundry Examples
Client configuration
If running Foundry as an MCP server via npx (when a compatible wrapper is available), use this config. Otherwise run the Python server directly.
{
"mcpServers": {
"foundry": {
"command": "python",
"args": ["-m", "promptise.server"],
"env": {
"OPENAI_API_KEY": "sk-..."
}
}
}
}Prompts to try
Use these prompts when interacting with a Foundry-powered agent.
- "Discover all available MCP tools and list their capabilities."
- "Research the latest developments in Model Context Protocol and write a summary."
- "Remember that our project deadline is June 30th and the team lead is Alice."
- "Run the ReAct reasoning pattern to solve: what is the fastest way to process a 10GB CSV file in Python?"Troubleshooting Foundry
pip install promptise fails with a dependency conflict
Create a fresh virtual environment ('python -m venv .venv && source .venv/bin/activate') before installing to avoid conflicts with globally installed packages.
Agent fails to discover MCP tools at startup
Verify that the MCP server command in the mcp_servers list runs successfully as a standalone process. Ensure Node.js is installed if using npx-based servers.
ChromaProvider raises a persistence error on startup
Make sure the persist_directory path exists and is writable. Install chromadb separately with 'pip install chromadb' if it was not pulled in as a dependency.
Frequently Asked Questions about Foundry
What is Foundry?
Foundry is a Model Context Protocol (MCP) server that foundation layer for agentic intelligence. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Foundry?
Follow the installation instructions on the Foundry GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Foundry?
Foundry works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Foundry free to use?
Yes, Foundry is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Foundry Alternatives — Similar Coding Agents Servers
Looking for alternatives to Foundry? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Foundry in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Foundry?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.