# OpenPitch

Gives coding agents sourced, confidence-scored intelligence on AI startups. Use it to check company metrics and compare conflicting public reports.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 11
- Author: Avierovich
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/Avierovich/openpitch
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/mcp/openpitch

## What it does

OpenPitch lets an agent query a public-source dataset about AI startups. Its MCP server reads committed records and returns metrics with provenance, confidence scores, and change history. When reports disagree, the result can include a consensus range and a contradiction flag.

The project processes podcasts, news, SEC EDGAR filings, and web material. A daily pipeline selects sources, transcribes content, extracts figures, reconciles conflicts, and publishes updates to a git repository.

The server is read-only and starts with `uvx openpitch-mcp`; no API key or signup is required. Treat its figures as probabilistic intelligence: each value carries a source, date, and confidence score, while confidence falls as the data gets older.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add openpitch -- uvx openpitch-mcp
```
uvx runs the command named after the package; if the README names another one, use uvx --from openpitch-mcp <command>

### Codex

**Platform CLI**
```bash
codex mcp add openpitch -- uvx openpitch-mcp
```
uvx runs the command named after the package; if the README names another one, use uvx --from openpitch-mcp <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "openpitch": {
            "command": "uvx",
            "args": [
                "openpitch-mcp"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from openpitch-mcp <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "openpitch": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "openpitch-mcp"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from openpitch-mcp <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add openpitch uvx openpitch-mcp
```
uvx runs the command named after the package; if the README names another one, use uvx --from openpitch-mcp <command>

## When to use it
- look up an AI startup’s ARR or valuation
- verify a company metric against public sources
- compare conflicting reports about one company
- track how a reported metric changed

## Example prompts
- Find Cognition’s ARR and show the sources plus confidence score.
- Check Sierra’s valuation and flag disagreements among public reports.
- Use `get_metric` and `get_provenance` to explain the latest changes for an AI startup.

## Similar
- [langchain-rag](https://affpapa.org/en/nevibecode/skills/langchain-rag.md) — skill · 17500 installs
- [marimo-pair](https://affpapa.org/en/nevibecode/skills/marimo-pair.md) — skill · 12700 installs
- [trace-to-training-data](https://affpapa.org/en/nevibecode/skills/trace-to-training-data.md) — skill · 2600 installs
- [AgentDB Advanced Features](https://affpapa.org/en/nevibecode/skills/agentdb-advanced-features.md) — skill · 1100 installs
- [ReasoningBank with AgentDB](https://affpapa.org/en/nevibecode/skills/reasoningbank-with-agentdb.md) — skill · 1100 installs
- [rag-engineer](https://affpapa.org/en/nevibecode/skills/sickn33-rag-engineer.md) — skill · 1000 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md