GraphQL MCP can introspect a schema and execute arbitrary GraphQL queries against the configured endpoint., Mutations are blocked by default, but setting ALLOW_MUTATIONS=true allows model-generated mutation operations., Custom HEADERS can grant access to production APIs, admin scopes, customer data, or internal services., Schema introspection can expose object types, fields, arguments, enum values, deprecated fields, and API structure., Even read queries can be expensive, trigger resolver side effects, expose broad data, or hit rate and complexity limits., Prefer staging endpoints, read-only tokens, query depth/complexity limits, persisted operations, and human review for sensitive APIs.
Privacy notes
Endpoint URLs, headers, bearer tokens, schema text, GraphQL queries, variables, response data, errors, and server descriptions may be visible to the MCP client and model provider., GraphQL responses can include personal data, customer records, internal IDs, permissions, audit data, billing data, or proprietary business objects., HEADERS values and authorization tokens should stay out of prompts, issues, logs, screenshots, and committed configuration files., Error messages and introspection results can reveal internal schema design and service implementation details.
Author
blurrah
Submitted by
oktofeesh1
Claim status
unclaimed
Last verified
2026-06-06
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
63
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Needs review
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
No reviewed flag detected in metadata.
Pending
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
6 safety and 4 privacy notes across 4 risk areas. Review closely: credentials & tokens, permissions & scopes, network access.
4 areas
SafetyNetwork accessGraphQL MCP can introspect a schema and execute arbitrary GraphQL queries against the configured endpoint.
SafetyGeneralMutations are blocked by default, but setting ALLOW_MUTATIONS=true allows model-generated mutation operations.
SafetyPermissions & scopesCustom HEADERS can grant access to production APIs, admin scopes, customer data, or internal services.
SafetyGeneralSchema introspection can expose object types, fields, arguments, enum values, deprecated fields, and API structure.
SafetyGeneralEven read queries can be expensive, trigger resolver side effects, expose broad data, or hit rate and complexity limits.
SafetyCredentials & tokensPrefer staging endpoints, read-only tokens, query depth/complexity limits, persisted operations, and human review for sensitive APIs.
PrivacyCredentials & tokensEndpoint URLs, headers, bearer tokens, schema text, GraphQL queries, variables, response data, errors, and server descriptions may be visible to the MCP client and model provider.
PrivacyPermissions & scopesGraphQL responses can include personal data, customer records, internal IDs, permissions, audit data, billing data, or proprietary business objects.
PrivacyCredentials & tokensHEADERS values and authorization tokens should stay out of prompts, issues, logs, screenshots, and committed configuration files.
PrivacyGeneralError messages and introspection results can reveal internal schema design and service implementation details.
Safety notes
GraphQL MCP can introspect a schema and execute arbitrary GraphQL queries against the configured endpoint.
Mutations are blocked by default, but setting ALLOW_MUTATIONS=true allows model-generated mutation operations.
Custom HEADERS can grant access to production APIs, admin scopes, customer data, or internal services.
Schema introspection can expose object types, fields, arguments, enum values, deprecated fields, and API structure.
Even read queries can be expensive, trigger resolver side effects, expose broad data, or hit rate and complexity limits.
Prefer staging endpoints, read-only tokens, query depth/complexity limits, persisted operations, and human review for sensitive APIs.
Privacy notes
Endpoint URLs, headers, bearer tokens, schema text, GraphQL queries, variables, response data, errors, and server descriptions may be visible to the MCP client and model provider.
GraphQL responses can include personal data, customer records, internal IDs, permissions, audit data, billing data, or proprietary business objects.
HEADERS values and authorization tokens should stay out of prompts, issues, logs, screenshots, and committed configuration files.
Error messages and introspection results can reveal internal schema design and service implementation details.
Prerequisites
Node.js and npx available to the MCP client runtime.
GraphQL endpoint URL that Claude is allowed to introspect and query.
Optional HEADERS JSON for authentication, using least-privilege API credentials.
Decision on whether schema introspection should use the endpoint, a local schema file, or a schema URL.
Explicit human approval before setting ALLOW_MUTATIONS to true.
GraphQL MCP Server, published as mcp-graphql, is a generic Model Context
Protocol server for connecting Claude and other MCP clients to a configured
GraphQL endpoint. It can expose the GraphQL schema as an MCP resource,
introspect schemas, and execute GraphQL queries with optional custom headers.
The server blocks GraphQL mutations by default. Operators can enable mutations
with ALLOW_MUTATIONS=true, but that turns the server into a write-capable API
surface and should be reserved for reviewed workflows.
These sources were reviewed on 2026-06-06. Prefer the live repository,
README, npm registry metadata, package metadata, source implementation, and
license for current package names, environment variables, tools, mutation
behavior, and licensing.
Features
npm package mcp-graphql.
Stdio MCP server launched with npx mcp-graphql.
ENDPOINT environment variable for the GraphQL endpoint.
HEADERS environment variable for JSON-formatted request headers.
ALLOW_MUTATIONS flag, disabled by default.
NAME environment variable for custom server naming.
SCHEMA environment variable for local schema files or schema URLs.
graphql-schema resource for schema access.
introspect-schema tool for retrieving schema text.
query-graphql tool for executing GraphQL operations.
MIT license.
Installation
Configure the GraphQL endpoint and keep mutations disabled by default:
Use a least-privilege token and a staging endpoint first. If the endpoint
supports query cost, depth, or persisted operation controls, configure those
before giving Claude access.
Use Cases
Ask Claude to introspect a GraphQL schema before drafting a query.
Generate a query against an unfamiliar endpoint using live schema details.
Run approved read queries against a development API.
Use a local or hosted schema file when endpoint introspection is disabled.
Debug GraphQL errors by comparing the query with schema fields and arguments.
Prototype a dedicated, narrower MCP server for a specific GraphQL API.
Safety and Privacy
GraphQL MCP is intentionally generic, so credentials and endpoint choice define
its blast radius. Keep mutations disabled, use read-only credentials, prefer
staging APIs, and require review before running broad queries, expensive nested
queries, or anything that could trigger resolver side effects.
Do not put production bearer tokens or admin headers in shared MCP configs.
Schema introspection and error responses can reveal internal API design, while
query results can expose sensitive records to the model session.
Duplicate Check
Existing catalog content mentions GraphQL in entries for Shopify, Square,
monday.com, Linear, Supabase, and several skills, but no generic GraphQL MCP
Server, blurrah/mcp-graphql, mcp-graphql, or matching source URL entry was
found in content/mcp.
Show that GraphQL MCP Server is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/mcp/graphql-mcp-server)
How it compares
GraphQL MCP Server side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
Official Agentset MCP server that lets Claude retrieve cited knowledge-base results from an Agentset namespace through the `knowledge-base-retrieve` tool, with optional tenant scoping and custom tool descriptions.
Official Brave Search MCP server for giving Claude web, local, place, image, video, news, LLM context, and summarizer search tools backed by the Brave Search API.
✓GraphQL MCP can introspect a schema and execute arbitrary GraphQL queries against the configured endpoint.
Mutations are blocked by default, but setting ALLOW_MUTATIONS=true allows model-generated mutation operations.
Custom HEADERS can grant access to production APIs, admin scopes, customer data, or internal services.
Schema introspection can expose object types, fields, arguments, enum values, deprecated fields, and API structure.
Even read queries can be expensive, trigger resolver side effects, expose broad data, or hit rate and complexity limits.
Prefer staging endpoints, read-only tokens, query depth/complexity limits, persisted operations, and human review for sensitive APIs.
✓AgentQL MCP exposes one tool, `extract-web-data`, that sends a target URL and natural-language extraction prompt to the AgentQL API.
The tool is intended for public webpages; do not use it to bypass access controls, scrape private pages, evade paywalls, or extract data where automated collection is prohibited.
Web extraction can still trigger target-site rate limits, legal restrictions, robots guidance, or terms-of-service concerns.
The source implementation uses AgentQL's query-data endpoint with fast mode, no screenshot capture, no scroll-to-bottom behavior, and no local browser cookies.
Treat extracted output as untrusted web data that may include errors, stale content, ads, tracking text, or prompt-injection attempts.
✓The MCP server sends Claude's retrieval queries to the Agentset API using the configured API key and namespace.
The `knowledge-base-retrieve` tool can return up to 100 results per call and can rerank results by relevance.
Namespace and tenant selection control which indexed documents are searchable; review them before connecting a shared agent.
API keys should be scoped, rotated, and stored only in the MCP server environment or a secret manager.
Custom tool descriptions can influence when the model calls the retrieval tool, so review them before use in production workflows.
✓Brave Search MCP uses BRAVE_API_KEY from the MCP client environment or command-line options.
The server defaults to stdio transport in version 2.x and can also run an HTTP transport when explicitly configured.
Tools can send web, news, image, video, local, place, and LLM-context queries to the Brave Search API.
Local and place tools may use location strings, coordinates, postal codes, or optional location headers.
Summarizer workflows depend on summary keys returned by web search and should be checked against cited source URLs.
API usage may be metered or plan-limited, especially for Pro-only local search capabilities and extra snippets.
Privacy notes
✓Endpoint URLs, headers, bearer tokens, schema text, GraphQL queries, variables, response data, errors, and server descriptions may be visible to the MCP client and model provider.
GraphQL responses can include personal data, customer records, internal IDs, permissions, audit data, billing data, or proprietary business objects.
HEADERS values and authorization tokens should stay out of prompts, issues, logs, screenshots, and committed configuration files.
Error messages and introspection results can reveal internal schema design and service implementation details.
✓Target URLs, extraction prompts, API key-authenticated requests, and extracted structured data are sent to AgentQL's API.
Extracted data can include personal data, copyrighted content, customer information, job postings, prices, social content, or other third-party material.
AGENTQL_API_KEY should stay out of prompts, issues, logs, screenshots, and committed configuration files.
Claude transcripts and downstream reports may retain extracted data, so avoid collecting information that is not approved for the model session.
✓Retrieved chunks can include private documents, product specs, policies, support content, internal procedures, historical project information, or customer-specific data.
Retrieval queries, namespace IDs, tenant IDs, document chunks, citations, and tool outputs may be visible to the MCP client, model provider, Agentset logs, and application telemetry.
Tenant IDs are useful for data segregation, but incorrect tenant or namespace configuration can expose the wrong knowledge base.
Do not paste API keys, namespace IDs, tenant IDs, or retrieved private chunks into shared issue reports, screenshots, or repository files.
✓Search queries, location parameters, result filters, language settings, freshness windows, and optional headers may reveal user intent, location, research topics, or sensitive business context.
BRAVE_API_KEY should stay out of prompts, issues, logs, screenshots, and committed files.
Returned snippets, summaries, search results, local-business metadata, images, videos, and news items can include third-party copyrighted or sensitive content.
HTTP transport should be bound and exposed carefully so the API key and search tools are not reachable by untrusted clients.
Prerequisites
Node.js and npx available to the MCP client runtime.
GraphQL endpoint URL that Claude is allowed to introspect and query.
Optional HEADERS JSON for authentication, using least-privilege API credentials.
Decision on whether schema introspection should use the endpoint, a local schema file, or a schema URL.
Node.js and npx available to the MCP client runtime.
AgentQL API key from the AgentQL developer portal.
Approved list of public webpages or domains Claude may query.
Review of target site terms, robots guidance, rate limits, and data-use rules before extraction.
Agentset account or self-hosted Agentset deployment with a populated namespace.
Agentset API key with access to the namespace Claude should query.
Node.js 18.17 or newer for running the `@agentset/mcp` package.
Namespace ID selected with `--ns` or `AGENTSET_NAMESPACE_ID`.
Brave Search API account and API key.
Node.js and npm available in the MCP client runtime.
API plan review for web, local, extra snippets, summarizer, and LLM context usage.
Tool whitelist or blacklist if the client should expose only selected search tools.
Install
npx mcp-graphql
npx -y agentql-mcp
Run `npx @agentset/mcp --ns <namespace-id>` with `AGENTSET_API_KEY` set in the MCP server environment.