MCP.so
Sign In
O

OpenPitch

@Avierovich

Repository

About OpenPitch

Open, real-time intelligence on AI startups — valuations, ARR, funding — every figure sourced, dated, and confidence-scored, with public-source contradiction flags. A free PitchBook alternative your AI agent can read: read-only, zero LLM calls, no API key. Run with: uvx openpitch

Basic information

Category

AI & Agents

Transports

stdio

Publisher

Avierovich

Submitted by

Avierovich

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "openpitch": {
      "command": "uvx",
      "args": [
        "openpitch-mcp"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is OpenPitch?

OpenPitch is a free, open-source MCP server that provides a real-time intelligence layer for AI startups — acting as a public, confidence-scored, sourced alternative to PitchBook and CB Insights. It mines podcasts, news, SEC filings, and web sources to extract metrics like valuation, ARR, and funding, then reconciles contradictions into a consensus with provenance. It's built for AI agents (Claude Code, Codex) and for anyone who needs a fresh, traceable first look at AI companies.

How to use OpenPitch?

No API key or signup required. Use uvx openpitch-mcp for zero-install, or pip install openpitch then run openpitch-mcp. Configure your agent's MCP settings as shown in the README. For the pipeline (rebuild data), clone the repo and use openpitch seed. Data lives in the git repo as committed JSON, read by the local server.

Key features of OpenPitch

  • Mines podcasts, news, SEC filings, and the web for metrics
  • Every figure linked to its original source (podcast timestamp, filing, article)
  • Confidence-scored based on source reliability, corroboration, and freshness
  • Reconciles conflicting sources, surfacing consensus range and contradiction flags
  • Learns which sources to trust over time
  • Version-tracked via git history as audit log
  • Composable: emits typed events other agents can subscribe to
  • A2A-discoverable via generated agent card
  • 60-second install, free to run and use

Use cases of OpenPitch

  • Ask your coding agent for AI startup metrics with sourced confidence
  • Build a newsletter agent that triggers on material events
  • Create a press/PR workflow that routes funding/valuation events above a confidence threshold
  • Drive investor outbound by monitoring universe entries and growth thresholds
  • Give your AI a grounded, sourced fact base to avoid hallucinating AI‑company numbers

FAQ from OpenPitch

Is OpenPitch really free? No API key or signup?

Yes — the data is committed to a public repository. The MCP server reads it locally. There is no key, no signup, and it runs entirely on free tiers. No cost to run or use.

What data does OpenPitch cover?

Global AI startups (140+ profiled across 12 sectors) and a dedicated MENA AI/tech segment. The top 50 are dynamically ranked by VC attention. Coverage grows daily via auto‑discovery.

How does OpenPitch handle conflicting sources?

When sources disagree, OpenPitch returns a consensus range plus a contradiction flag. It never silently guesses — you see the variance and can inspect each source.

What transport does the MCP server use?

The MCP server runs locally over stdio. Configuration is done via the agent's MCP config (e.g., Claude Code or Codex). There is no server‑side component.

What are the known limits of OpenPitch?

Figures are probabilistic, not verified. Coverage is narrow by design (global AI + MENA segment). It is not investment advice. Always verify before acting. Open quality items are tracked publicly.

Comments

More AI & Agents MCP servers

Just Prompt - A lightweight MCP server for LLM providers

disler

just-prompt is an MCP server that provides a unified interface to top LLM providers (OpenAI, Anthropic, Google Gemini, Groq, DeepSeek, and Ollama)

733

Model Context Protocol for Unreal Engine

chongdashu

Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).

2.0K

🛡️ A.I.G（AI-Infra-Guard）

Tencent

A full-stack AI Red Teaming platform securing AI ecosystems via OpenClaw Security Scan, Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation.

4.1K

欢迎来到 智言平台

Shy2593666979

AgentChat 是一个基于 LLM 的智能体交流平台，内置默认 Agent 并支持用户自定义 Agent。通过多轮对话和任务协作，Agent 可以理解并协助完成复杂任务。项目集成 LangChain、Function Call、MCP 协议、RAG、Memory、HITL、Skill、Milvus 和 ElasticSearch 等技术，实现高效的知识检索与工具调用，使用 FastAPI 构建高性能后端服务。

797

MCP-LLM Bridge

patruff

Bridge between Ollama and MCP servers, enabling local LLMs to use Model Context Protocol tools

976