introspect-schema
Introspect the GraphQL schema, use this tool before doing a query to get the schema information if you do not have it available as a resource already.
Give a model a GraphQL endpoint it has never seen: introspect the schema first, then run queries with variables against the same endpoint.
LABSCO SUMMARY
The order carries the value: introspect-schema first so the types and fields are known, then query-graphql with the document and its variables. There are no per-operation tools and no query builder — one call sends whatever you pass — so the schema is what constrains the model, and whatever guards the endpoint sits outside this server.
Two tools over one GraphQL endpoint — schema introspection, and query execution.
A GraphQL endpoint the server can reach, and a Node runtime. The server itself takes no credentials.
One command — npx -y mcp-graphql
USE CASES
Introspect the schema first so the types, fields and arguments are known, then send a query built against what came back rather than against a guess.
Introspection answers from the endpoint itself, so a field that is missing there is missing in the deployment, whatever the branch says.
A single query with its variables, without opening a client or writing a fetch — useful when the question is what the API returns for this ID, right now.
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
introspect-schemaIntrospect the GraphQL schema, use this tool before doing a query to get the schema information if you do not have it available as a resource already.
query-graphqlQuery a GraphQL endpoint with the given query and variables
BEFORE YOU INSTALL
2 tools when we connected.
npx on PATH
CONTEXT COST
Rated by the size of the 2 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Add the block to your client, or run the command directly.
{
"mcpServers": {
"mcp-graphql": {
"command": "npx",
"args": [
"-y",
"mcp-graphql"
]
}
}
}npx -y mcp-graphql