Meta Business MCP
A safe, self-hosted gateway for inspecting Meta Business assets and creating reviewable proposals without exposing broad Graph API access.
README
Meta Business MCP
Meta Business MCP is an open-source-ready skeleton for an auditable Model Context Protocol gateway over Meta Business APIs.
The goal is not to build an autonomous marketing agent. The goal is a safe, self-hosted operations gateway that can eventually let MCP clients inspect Meta Business assets, normalize performance data, and create reviewable proposals without exposing broad Graph API access.
This project is not affiliated with, endorsed by, or sponsored by Meta or OpenAI.
Current status
This repository is a public skeleton. It intentionally does not call live Meta APIs yet.
Included:
- TypeScript workspace with strict builds.
- Initial MCP tool contracts.
- Real stdio MCP server entrypoint.
- Default-safe policy primitives.
- Allowlisted Meta API boundary.
- Proposal hashing and state-transition primitives.
- Audit redaction helpers.
- Docs, examples, and tests.
Excluded from v1:
- Meta writes.
- Posting, messaging, WhatsApp, deletes, budget changes, or campaign activation.
- Raw lead export or lead PII retrieval.
- Generic
graph_request,meta_api_call, arbitrary URL fetch, or browser automation. - Clara-specific prompts, account IDs, workflows, thresholds, CRM integrations, or private playbooks.
Tool contracts
Initial contracts:
meta_connection_statusmeta_ad_accounts_listmeta_campaigns_listmeta_ads_insights_querymeta_pages_listmeta_instagram_accounts_listsearchfetchmeta_proposal_create_budget_changemeta_proposal_create_delivery_status_change
Every handler currently returns a safe skeleton response until an allowlisted Meta operation and mocked contract tests exist.
Development
npm install
npm run typecheck
npm test
npm run build
Run the stdio server skeleton:
npm --workspace @meta-business-mcp/server run dev:stdio
Install it in Codex:
codex mcp add meta-business -- \
npm --prefix /absolute/path/to/meta-business-mcp \
--workspace @meta-business-mcp/server \
run dev:stdio
See docs/install.md for Codex and ChatGPT setup notes.
Security defaults
- Read-only by default.
- No raw Meta token in tool arguments or outputs.
- No generic Graph proxy.
- No live credentials required for tests.
- No lead values in the skeleton.
- No Meta mutation methods in the allowlisted client boundary.
See docs/threat-model.md and docs/architecture.md.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.