Compatible with every major AI agent and IDE
What is the Convex MCP Server?
Connect your Convex deployment to any AI agent and manage your application's data and logic through natural conversation. This server allows you to interact with your real-time database and serverless functions without leaving your AI interface.
What you can do
- Data Fetching — Execute read-only queries to retrieve documents and state from your Convex tables.
- Transactional Updates — Run mutations to modify data with full ACID guarantees directly from the agent.
- Side Effects & APIs — Trigger Convex actions for external API calls, heavy computation, or non-transactional logic.
- Flexible Execution — Call functions using standard colon notation or URL-style identifiers for maximum compatibility.
How it works
- Subscribe to this server
- Enter your Convex Deployment URL (and optional Access Key)
- Start querying and mutating your data from Claude, Cursor, or any MCP-compatible client
Who is this for?
- Full-stack Developers — Debug data, run migrations, or check state directly from your IDE or chat.
- Product Managers — Query live application metrics and user data using natural language without writing code.
- Support Teams — Inspect and update user records or trigger administrative actions through a secure AI interface.
Built-in capabilities (4)
Call a Convex action function
g., "messages/list" instead of "messages:list"). Call a Convex function by its URL identifier
Call a Convex mutation function
Use this for fetching data. Call a Convex query function
Why AutoGen?
AutoGen enables multi-agent conversations where agents negotiate, delegate, and collaboratively use Convex tools. Connect 4 tools through Vinkius and assign role-based access. a data analyst queries while a reviewer validates, with optional human-in-the-loop approval for sensitive operations.
- —
Multi-agent conversations: multiple AutoGen agents discuss, delegate, and collaboratively use Convex tools to solve complex tasks
- —
Role-based architecture lets you assign Convex tool access to specific agents. a data analyst queries while a reviewer validates
- —
Human-in-the-loop support: agents can pause for human approval before executing sensitive Convex tool calls
- —
Code execution sandbox: AutoGen agents can write and run code that processes Convex tool responses in an isolated environment
Convex in AutoGen
Convex and 4,000+ other MCP servers. One platform. One governance layer.
Teams that connect Convex to AutoGen through Vinkius don't need to source, host, or maintain individual MCP servers. Every tool call runs inside a hardened runtime with credential isolation, DLP, and a signed audit chain.
Raw MCP | Vinkius | |
|---|---|---|
| Server catalog | Find and host yourself | 4,000+ managed |
| Infrastructure | Self-hosted | Sandboxed V8 isolates |
| Credential handling | Plaintext in config | Vault + runtime injection |
| Data loss prevention | None | Configurable DLP policies |
| Kill switch | None | Global instant shutdown |
| Financial circuit breakers | None | Per-server limits + alerts |
| Audit trail | None | Ed25519 signed logs |
| SIEM log streaming | None | Splunk, Datadog, Webhook |
| Honeytokens | None | Canary alerts on leak |
| Custom domains | Not applicable | DNS challenge verified |
| GDPR compliance | Manual effort | Automated purge + export |
Why teams choose Vinkius for Convex in AutoGen
The Convex MCP Server runs on Vinkius-managed infrastructure inside AWS — a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts. All 4 tools execute in hardened sandboxes optimized for native MCP execution.
Your AI agents in AutoGen only access the data you authorize, with DLP that blocks sensitive information from ever reaching the model, kill switch for instant shutdown, and up to 60% token savings. Enterprise-grade infrastructure, zero maintenance.

* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure
How Vinkius secures
Convex for AutoGen
Every tool call from AutoGen to the Convex MCP Server is protected by DLP redaction, cryptographic audit chains, V8 sandbox isolation, kill switch, and financial circuit breakers.
Frequently asked questions
What is the difference between run_query and run_mutation?
Use run_query for read-only operations that fetch data. Use run_mutation when you need to write, update, or delete data transactionally in your database.
Can I call external APIs using this server?
Yes, by using the run_action tool. Actions in Convex are designed for side effects like calling third-party APIs or performing long-running tasks.
How do I reference functions in subdirectories?
You can use run_query with colon notation (e.g., 'folder/file:function') or use run_function which accepts URL-style identifiers with slashes.
How does AutoGen connect to MCP servers?
Create an MCP tool adapter and assign it to one or more agents in the group chat. AutoGen agents can then call Convex tools during their conversation turns.
Can different agents have different MCP tool access?
Yes. AutoGen's role-based architecture lets you assign specific MCP tools to specific agents, so a querying agent has different capabilities than a reviewing agent.
Does AutoGen support human approval for tool calls?
Yes. Configure human-in-the-loop mode so agents pause and request approval before executing sensitive MCP tool calls.
McpWorkbench not found
Install: pip install "autogen-ext[mcp]"
Explore More MCP Servers
View all →
Ayuntamiento de Barcelona (CKAN)
7 toolsAccess the Open Data BCN portal to query urban datasets, environmental records, and public statistics directly through AI.

Stripe Legacy
11 toolsManage payments, customers, and account balances on Stripe via the legacy API with AI agents.

FreshBooks
12 toolsManage small business accounting via FreshBooks — track clients and invoices, handle payments and billing via AI agents.

Google Books
8 toolsSearch and explore millions of books on Google Books — find titles, authors, reviews, previews and full text from any AI agent.
