mcp-server

vouch: Controlled AI Knowledge

Name: vouch: Controlled AI Knowledge
Author: vouchdev

Get a review-gated knowledge base for AI agents with vouch. For founders using AI-powered tools.

From the repo · “A git-native, review-gated knowledge base for AI agents: they propose writes, you approve them. Every claim cites a source, every change is a diff in your repo. MCP + CLI.”

67 stars58 forksPythonGuide quality 8/10 · reviewed Jul 17Updated 8/23/2026100% free · open source

What it does

Vouch creates a review-gated knowledge base for AI agents, allowing them to propose changes that founders can approve, with every claim citing a source and every change tracked in a git repository.

When to use it

•When using AI-powered tools to automate decision-making processes
•When needing a transparent and accountable knowledge base for AI-driven insights
•When collaborating with team members to review and validate AI-generated content

Ready-to-paste prompt

vouch propose 'The sky is blue because of Rayleigh scattering.' --source 'https://en.wikipedia.org/wiki/Rayleigh_scattering'

Heads up: Vouch requires a git repository to track changes, so make sure you have git installed and configured on your system before using vouch.

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

✅ Light setup

Installs with a command or two; your AI agent can do it for you.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/vouch && curl -fsSL https://workflowstacks.com/api/skills/vouch/claude-skill -o ~/.claude/skills/vouch/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How vouch: Controlled AI Knowledge works

Codeflow

Free to inspect

Vouch: Controlled AI Knowledge is a large Python project (~58k lines across 248 code files, plus 153 test files). Setup is light: runs with Docker. Last commit 2 months ago, MIT license, has a test suite.

Size

Large codebase

~58k lines · 248 code files · ~8 h to skim

Setup

Light setup

Runs with Docker.

Runs on

Python · Node.js · Docker

No API keys detected

Python 88%TypeScript 10%Shell 1%CSS 1%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
webapp/app.py
Where the program starts running
4
package.json
Dependencies and the commands it exposes
5
src/vouch/__init__.py
Inside src/ — the main logic begins here

What's in each folder

src/Core code — the actual logic136 files

skills/Prompts, skills & agent definitions2 files

adapters/Sub-packages / plugins64 files

examples/Examples you can copy140 files

docs/Documentation67 files

templates/Templates10 files

demo/Examples you can copy6 files

webapp/Folder90 files

READMEHas testsDocumentedCI checksDocker readyExamples includedMIT licenseUpdated 2 mo ago

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

✅ Light setup

Installs with a command or two; your AI agent can do it for you.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/vouch && curl -fsSL https://workflowstacks.com/api/skills/vouch/claude-skill -o ~/.claude/skills/vouch/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

vouchdev

Language

Python

Do this with a working workflow

Pages we wrote and tested that put a skill like this to work — importable n8n templates, outcome guides and Claude Desktop configs.

MCP config

Memory (knowledge graph) MCP config

Persistent memory across chats via a local knowledge graph.

Ready-to-run pack · one-time $29

Review Watchdog

Catch every bad review the day it lands — with a reply ready.

Includes the tested workflow and a written setup playbook. Needs: Google Places API key (free tier).

See what you get

Related skills

More mcp-server tools founders pair with this one.

mcp-server★ 362,094

Rea: Reverse Engineer Apps

Inspect apps down to binary level with Rea. For founders needing code clarity.

mcp-server★ 156,432

open-webui: Self‑hosted AI Chat UI for Faster LLM Integration

Launch a sleek, web‑based AI interface that works with Ollama, OpenAI API, and more—ideal for founders needing a quick, customizable LLM front‑end. 153k+ GitHub stars.

mcp-server★ 132,268

graphify

Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.

mcp-server★ 143

logic-pro-mcp

Local MCP server for stateful, fail-closed Logic Pro control and live project readback.

mcp-server★ 398

mcpproxy-go: Secure AI Agents

Supercharge AI agents safely with mcpproxy-go, for founders using AI tools.

mcp-server★ 36,218

oh‑my‑pi: Supercharge Terminal Coding with an AI Assistant

Accelerate terminal development with hash‑anchored edits, LSP, Python and browser tools, plus multi‑LLM support. Made for founders coding in the CLI. 34k+ GitHub stars.