Hypertune MCP by usefulapi

by usefulapi.io · remote · no sign-in to list tools · v1.16.4 · published by io.usefulapi

Evaluate and inspect your Hypertune feature flags from your AI assistant.

Changes · 3

Compare Registry entry

5 tools: 3 change things and 2 only read.

Scanned Oct 22, 2026

Deletes or cancels0Claude Code: deny
Can't be undone0Claude Code: ask
Changes things3Claude Code: ask: see these tools
Only reads2Claude Code: allow: see these tools

3 checks matched something in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http hypertune https://hypertune.usefulapi.io/mcp

Findings

Matches of fixed checks, with the text they matched.

3 warn

Warnhypertune_evaluate
The server marks "hypertune_evaluate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
Matched text
annotations: {"readOnlyHint":true}; The name "hypertune_evaluate" says "evaluate": a word for a change.
What this check looks for
Warnhypertune_query
The server marks "hypertune_query" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
Matched text
annotations: {"readOnlyHint":true}; The name "hypertune_query" says "query": a word for reading.
What this check looks for
Warnhypertune_introspect
The server marks "hypertune_introspect" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
Matched text
annotations: {"readOnlyHint":true}; The description's first verb is "run": a word for a change.
What this check looks for

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

5 tools

Changes things 3Creates or changes something that can be changed back.Claude Code: ask

hypertune_evaluateEvaluate one or more feature flags for a given context and get their resolved values.1 finding
Why it is in this group
- The name "hypertune_evaluate" says "evaluate": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
  The server marks "hypertune_evaluate" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
  Matched text
  annotations: {"readOnlyHint":true}; The name "hypertune_evaluate" says "evaluate": a word for a change.
  What this check looks for
Description, as the server wrote it
Evaluate one or more feature flags for a given context and get their resolved values. Builds `query { root(context: <context>) { <selection> } }`. Provide `selection` = the flag fields to read (e.g. `exampleFlag anotherFlag { nestedField }`) and `context` = the evaluation context (environment + user/attributes). Every field must supply all of its arguments (Hypertune reduces the flag logic to a JSON result). Use hypertune_list_flags / hypertune_introspect first if you don't know the flag names.
Annotations the server declares
- title: Evaluate feature flags
- readOnlyHint: true
hypertune_introspectRun a GraphQL introspection query to discover the project's schema — all types and fields, including the available flags and the shape of the evaluation context.1 finding
Why it is in this group
- The description's first verb is "run": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
  The server marks "hypertune_introspect" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
  Matched text
  annotations: {"readOnlyHint":true}; The description's first verb is "run": a word for a change.
  What this check looks for
Description, as the server wrote it
Run a GraphQL introspection query to discover the project's schema — all types and fields, including the available flags and the shape of the evaluation context. Best-effort: some edge deployments disable introspection, in which case this returns a clear error (use hypertune_query with a known flag name instead).
Annotations the server declares
- title: Introspect the GraphQL schema
- readOnlyHint: true
hypertune_queryRun an arbitrary GraphQL query (and optional variables) against the Hypertune edge endpoint.1 finding
Why it is in this group
- The name "hypertune_query" says "query": a word for reading.
- The description's first verb is "run": a word for a change.
- The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
Findings
- Warn
  The server marks "hypertune_query" read-only (readOnlyHint: true), and its own name, description or input give the class "write".
  Matched text
  annotations: {"readOnlyHint":true}; The name "hypertune_query" says "query": a word for reading.
  What this check looks for
Description, as the server wrote it
Run an arbitrary GraphQL query (and optional variables) against the Hypertune edge endpoint. Power-user escape hatch for full control — each Hypertune project has its own generated schema, so this handles any flag shape, nested selections, inline enum literals, or variables the convenience tools can't express. The query root field is `root(context: ...)`.
Annotations the server declares
- title: Run a raw GraphQL query
- readOnlyHint: true

Only reads 2Looks things up and changes nothing.Claude Code: allow

hypertune_get_logicFetch the underlying LOGIC of flags without reducing them — i.e.No findings
Why it is in this group
- The name "hypertune_get_logic" says "get": a word for reading.
Description, as the server wrote it
Fetch the underlying LOGIC of flags without reducing them — i.e. `query { root { <selection> } }` with no context. Returns the flag's conditional expression / rules rather than a single resolved value, useful for inspecting how a flag is configured. If the schema requires a context arg you can still pass a partial one via `context`.
Annotations the server declares
- title: Get flag logic (unreduced)
- readOnlyHint: true
hypertune_list_flagsList the project's top-level feature flags (the fields of the query root type) with their return types and descriptions, derived from GraphQL introspection.No findings
Why it is in this group
- The name "hypertune_list_flags" says "list": a word for reading.
Description, as the server wrote it
List the project's top-level feature flags (the fields of the query root type) with their return types and descriptions, derived from GraphQL introspection. A convenient first call to see what flags exist before evaluating. Requires introspection to be enabled on the edge endpoint.
Annotations the server declares
- title: List feature flags
- readOnlyHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

No private-data, untrusted-input, outbound path in this server alone: no tool is tagged private-data or untrusted-input or outbound.

Reads private data0: Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write0: Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out0: Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Looked at Oct 22, 2026, 15:29 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable: yes
HTTP status of an anonymous initialize: 200
tools/list served without credentials: yes
Tools that change things, listed without credentials: 3
Protocol version: 2025-06-18
WWW-Authenticate on 401: not observed
resource_metadata in WWW-Authenticate: none
RFC 9728 protected resource metadata: https://hypertune.usefulapi.io/.well-known/oauth-protected-resource/mcp
Authorization servers: https://hypertune.usefulapi.io
Legacy HTTP+SSE transport: no
Dynamic client registration: yes
Client ID metadata documents: no

WarnThe whole server
The endpoint served tools/list without credentials, and the list has 3 tools that change things (3 write). Whether a call also works without credentials was not tested: tools are never called.
Matched text
https://hypertune.usefulapi.io/mcp: tools/list returned 5 tools to an anonymous client, including hypertune_evaluate (write), hypertune_query (write), hypertune_introspect (write)
What this check looks for

Install

Pinned to version 1.16.4. Values in angle brackets are yours to fill in.

claude mcp add --transport http hypertune https://hypertune.usefulapi.io/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json

{
  "permissions": {
    "allow": [
      "mcp__hypertune__hypertune_get_logic",
      "mcp__hypertune__hypertune_list_flags"
    ],
    "ask": [
      "mcp__hypertune__hypertune_evaluate",
      "mcp__hypertune__hypertune_query",
      "mcp__hypertune__hypertune_introspect"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as hypertune. Review it before you use it.

Keep track

Get told when this server's tools change: a new tool, a new description, a new class.

https://anlyon.dev/servers/io.usefulapi/hypertune/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Anlyon alternative

3 of these tools change something. Run those as governed actions.

Declare what an action changes, and Anlyon puts it under a shared limit, binds every approval to the exact request, and grades the outcome. It works on any HTTPS API you can describe as an action.

Anlyon does not run, proxy or govern this MCP server. A tool your own code calls directly is not routed through Anlyon and is not governed by it.

Import an OpenAPI spec →

Compare with other servers for the same product.

Source: MCP Registry · repository. Synced Oct 22, 2026, 11:18 UTC.

For agents: Markdown · JSON · report this listing