Skip to main content
Glama

Convex MCP server

Official
by get-convex
README.md1.5 kB
## About npm-packages contains the cli, demos, tutorials, tests, scripts and more. We use Rush for package management, so don't run `npm install` in these directories. See the main convex/README.md file for details. ## Adding a new convex package To add a new package that uses convex, follow these steps instead of our quickstart: 1. Create a new subdirectory 2. Copy the contents of the npm-packages/tutorial into the subdirectory 3. Update `"name"` in your new `package.json` file to match your subdirectory name 4. Add your new package to [rush](https://github.com/get-convex/convex/blob/main/npm-packages/rush.json#L296) 5. Run `just rush update` to install dependencies ## Running a convex package You can run projects against either a local backend, or against prod. If your project requires changes that have not yet been deployed, you'll need to test against a local backend. To use a local backend: 1. Run `just run-local-backend` 2. In a separate terminal window, run `just convex dev` To use a local big-brain (runs both backend and big brain locally): 1. Run `just run-big-brain` 2. Copy the `Test with:` line output by big-brain: `2023-07-12T19:07:07.422400Z INFO big_brain::model: Test with: CONVEX_PROVISION_HOST=http://0.0.0.0:8050 npx convex dev --override-auth-url "https://cheerful-lake-55-staging.authkit.app/" " --override-auth-client "XXXXXXXXX"` 3. Paste it into another terminal that's in your convex directory. To run against prod: Just run `npx convex dev`

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/get-convex/convex-backend'

If you have feedback or need assistance with the MCP directory API, please join our Discord server