Ermöglicht es KI-Assistenten, einen Katalog von über 160 generativen Modellen zu durchsuchen und Aufgaben für die Generierung von Bildern, Videos, Musik und Audio auszuführen.
Auch bekannt als runapi-ai/mcp, mcp, @runapi.ai/mcp
- Stars
- 56
- Tools
- 9
- Downloads, 30 Tage
- —
- Neueste Version
- 0.17.0
- Letzter Commit
- 10. Nov. 2026
- Lizenz
- Apache-2.0
Was ist RunAPI MCP?
Ermöglicht es KI-Assistenten, einen Katalog von über 160 generativen Modellen zu durchsuchen und Aufgaben für die Generierung von Bildern, Videos, Musik und Audio auszuführen. Er stellt 9 Tools bereit, etwa list_models, get_model_info, list_actions.
Zusammenfassung generiert aus dem README des Projekts
Über RunAPI MCP server
Der RunAPI MCP Server verbindet KI-Assistenten mit einem riesigen Katalog von über 160 generativen Modellen. Er ermöglicht es Entwicklern, mit der Medienproduktion in vier verschiedenen Modalitäten zu interagieren: Bild, Video, Audio und Utility. Nutzer können verfügbare Modelle durchsuchen, spezifische Input-Constraints prüfen und Echtzeit-Preise über Discovery-Tools abrufen, die keinen API-Key erfordern.
Für die aktive Medienproduktion bietet der Server authentifizierte Tools, um Tasks zu erstellen, deren Status abzufragen und das Account-Guthaben zu verwalten. Er fungiert als reiner Client, was bedeutet, dass er kein lokales Generation-Backend ausführt. Er ist kompatibel mit verschiedenen Coding-Umgebungen wie Claude Code, Cursor, Windsurf, VS Code und Roo Code. Nutzer können den Server lokal via npx einrichten oder sich über eine direkte URL mit einer remote gehosteten Version verbinden. Die Authentifizierung erfolgt über ein browserbasiertes Login-Tool, Umgebungsvariablen oder eine gemeinsame Config-Datei.
Was du damit machen kannst
- 1Account-Nutzung überwachenÜberprüfe deine aktuellen Spending-Metriken und das verbleibende Account-Guthaben.
Geschrieben aus dem README des Projekts. Jeder Use Case wurde gegen ein Zitat daraus geprüft.
So installierst du RunAPI MCP
Formate geprüft am 10. Nov. 2026Zu Cursor hinzufügenZu Goose hinzufügenRunAPI MCP in Claude Code installieren
$ claude mcp add --env 'RUNAPI_API_KEY=<YOUR_RUNAPI_API_KEY>' mcp -- npx -y @runapi.ai/mcpRunAPI MCP in Cursor installieren
~/.cursor/mcp.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@runapi.ai/mcp"
],
"env": {
"RUNAPI_API_KEY": "<YOUR_RUNAPI_API_KEY>"
}
}
}
}RunAPI MCP in VS Code installieren
.vscode/mcp.json
{
"servers": {
"mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@runapi.ai/mcp"
],
"env": {
"RUNAPI_API_KEY": "<YOUR_RUNAPI_API_KEY>"
}
}
}
}RunAPI MCP in VS Code CLI installieren
$ code --add-mcp '{"name":"mcp","type":"stdio","command":"npx","args":["-y","@runapi.ai/mcp"],"env":{"RUNAPI_API_KEY":"<YOUR_RUNAPI_API_KEY>"}}'RunAPI MCP in Claude Desktop installieren
claude_desktop_config.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@runapi.ai/mcp"
],
"env": {
"RUNAPI_API_KEY": "<YOUR_RUNAPI_API_KEY>"
}
}
}
}RunAPI MCP in Codex installieren
$ codex mcp add mcp --env 'RUNAPI_API_KEY=<YOUR_RUNAPI_API_KEY>' -- npx -y @runapi.ai/mcpRunAPI MCP in Codex config.toml installieren
~/.codex/config.toml
[mcp_servers.mcp] command = "npx" args = ["-y", "@runapi.ai/mcp"] [mcp_servers.mcp.env] RUNAPI_API_KEY = "<YOUR_RUNAPI_API_KEY>"
RunAPI MCP in Gemini CLI installieren
$ gemini mcp add -s user -e 'RUNAPI_API_KEY=<YOUR_RUNAPI_API_KEY>' mcp npx -- -y @runapi.ai/mcpRunAPI MCP in Gemini settings.json installieren
~/.gemini/settings.json
{
"mcpServers": {
"mcp": {
"command": "npx",
"args": [
"-y",
"@runapi.ai/mcp"
],
"env": {
"RUNAPI_API_KEY": "<YOUR_RUNAPI_API_KEY>"
}
}
}
}Von HostedMCP gestartet am 16. Nov. 2026
Wir haben diesen Befehl für @runapi.ai/mcp 0.17.0 in einem isolierten Container ohne Netzwerk und ohne Schlüssel ausgeführt. Er hat geantwortet und 9 Tools gelistet.
RunAPI MCP tools 9
Vom Server mit tools/list gelesenlist_modelsList RunAPI models from the embedded catalog. Optional filters: modality, service, or action.
get_model_infoGet supported endpoint, current runtime pricing, and input constraints for a RunAPI model slug. Add service and action when the model supports multiple endpoints.
list_actionsList RunAPI endpoint names grouped by output modality.
check_pricingReturn current runtime pricing for a RunAPI model/action pair.
search_promptsSearch RunAPI prompt examples by modality, category, tags, text query, or model. Each prompt includes its approved example images in media; example videos and reference images are not returned. Free, no API key required.
check_balanceReturn the authenticated RunAPI account balance and spending metrics.
create_taskRun a RunAPI operation with a caller-generated idempotency key. Asynchronous operations can optionally poll until completion.
get_taskFetch the current status and latest payload for an existing RunAPI task.
loginAuthenticate RunAPI by opening a browser PKCE login flow and saving the API key to ~/.config/runapi/config.json.
| list_models | List RunAPI models from the embedded catalog. Optional filters: modality, service, or action. | |
| get_model_info | Get supported endpoint, current runtime pricing, and input constraints for a RunAPI model slug. Add service and action when the model supports multiple endpoints. | |
| list_actions | List RunAPI endpoint names grouped by output modality. | |
| check_pricing | Return current runtime pricing for a RunAPI model/action pair. | |
| search_prompts | Search RunAPI prompt examples by modality, category, tags, text query, or model. Each prompt includes its approved example images in media; example videos and reference images are not returned. Free, no API key required. | |
| check_balance | Return the authenticated RunAPI account balance and spending metrics. | |
| create_task | Run a RunAPI operation with a caller-generated idempotency key. Asynchronous operations can optionally poll until completion. | |
| get_task | Fetch the current status and latest payload for an existing RunAPI task. | |
| login | Authenticate RunAPI by opening a browser PKCE login flow and saving the API key to ~/.config/runapi/config.json. |
Konfiguration
RUNAPI_API_KEYSecretOptional RunAPI API key for headless hosts. Interactive MCP hosts can use the login tool, which writes shared RunAPI config.
| Name | Beschreibung | Typ |
|---|---|---|
| RUNAPI_API_KEY | Optional RunAPI API key for headless hosts. Interactive MCP hosts can use the login tool, which writes shared RunAPI config. | Secret |
Automatisierte Checks
Geluften am 16. Nov. 2026Vollständiger Bericht →Unsere automatisierten Checks bestanden. Das ist kein Security-Audit: Lies den Code und gib dem Server nur den minimal notwendigen Zugriff. Wie die Checks funktionieren
- Source Repository verknüpft und erreichbarrunapi-ai/mcp is public and reachablegithub.com/runapi-ai/mcp ↗Bestanden
- Paket nennt dieses Repository als Quelle@runapi.ai/mcp names runapi-ai/mcp as sourcenpmjs.com/package/@runapi.ai/mcp ↗Bestanden
- Build-Provenienz (Sigstore-Attestierung)No build attestation published for this releaseNicht geprüft
- Open Source Lizenz (OSI)Apache-2.0spdx.org/licenses/Apache-2.0.html ↗Bestanden
- Commit oder Release im letzten JahrLast commit or release 2026-11-10github.com/runapi-ai/mcp ↗Bestanden
- OpenSSF ScorecardNicht geprüft
- Package-Name ähnelt keinem populären PackageNo popular package one edit away from @runapi.ai/mcpBestanden
- Keine kritischen Advisories im Paket oder seinen Abhängigkeiten100 resolved dependencies checked; no critical advisoryosv.dev ↗Bestanden
- Keine versteckten Anweisungen an das Modell9 tool descriptions and the README read; no override phrases, credential paths or hidden charactersBestanden
- Destruktive Tools sind markiertNo tool named like delete, drop or execBestanden
- Remote-Endpoint beantwortet den MCP-HandshakeHTTPS endpoint answered 401 (sign-in required)Bestanden
README
Auf GitHub ansehenWhat Is This?
RunAPI MCP Server connects MCP-compatible coding tools to RunAPI. It lets an assistant browse the RunAPI catalog, inspect model inputs, check current runtime pricing, create media tasks, poll task status, and check account balance.
The discovery tools work without an API key because they use the embedded build-time catalog.
Authenticated operations use the login tool, runapi login, RUNAPI_API_KEY, or shared RunAPI config.
This package is a pure client. It does not run a local generation backend and only needs RunAPI sign-in for authenticated tools.
Hosted MCP
Remote MCP clients can connect directly to:
https://mcp.runapi.ai/mcp
No Node.js installation is required. Use OAuth when the client supports remote MCP authorization, or configure a RunAPI API key as the bearer credential. See the Hosted MCP setup page for Cursor, Claude Desktop, VS Code, and Windsurf configurations.
Hosted MCP exposes the eight Business Tools documented below. Local Login remains available only in the Local MCP package.
Quick Start
For Local MCP over stdio, install RunAPI with Claude Code's MCP command:
claude mcp add runapi -s user -- npx -y @runapi.ai/mcp
The scope flag controls where the MCP server is stored:
-s user: global, available in all projects for your user.-s project: team-shared, written to.mcp.jsonin the current repo so it can be committed.
Use project scope when you want the whole team to share the same server config:
claude mcp add runapi -s project -- npx -y @runapi.ai/mcp
Compatibility fallback for non-Claude Code platforms or manual JSON config:
{
"mcpServers": {
"runapi": {
"command": "npx",
"args": ["-y", "@runapi.ai/mcp"]
}
}
}
If your host needs a generated config file, use the legacy init command as a fallback:
npx @runapi.ai/mcp init claude
npx @runapi.ai/mcp init cursor
npx @runapi.ai/mcp init vscode
npx @runapi.ai/mcp init windsurf
npx @runapi.ai/mcp init roo
Free catalog tools work before sign-in.
For task creation, task status, and balance checks, ask your assistant to call the login tool. It opens a browser login and saves credentials to ~/.config/runapi/config.json, the same file used by runapi login.
Headless and CI hosts can still set RUNAPI_API_KEY before starting the MCP host.
Tools
| Tool | Auth | Purpose |
|---|---|---|
list_models |
No | List RunAPI models from the embedded catalog. Supports modality, service, and action filters. |
get_model_info |
No | Return service, action, modality, input constraints, and current runtime pricing for a model slug. Use service + action when a model appears in multiple endpoints. |
list_actions |
No | Group endpoint action names by modality. |
check_pricing |
No | Return current runtime pricing for a service + action + model combination. |
search_prompts |
No | Search reusable prompt examples by modality, category, tags, q, model, featured, and pagination. Each prompt returns media: its approved example images, including the cover image of a video prompt. Example videos and reference images are not returned yet. |
login |
No | Start browser login and save RunAPI credentials to shared local config. |
create_task |
Yes | Create a media task with a required caller-generated idempotency_key; wait for completion by default or return immediately with wait: false. |
get_task |
Yes | Fetch status and latest payload for an existing media task. |
check_balance |
Yes | Return account balance and spending metrics. |
The catalog, pricing, and prompt search tools are designed for funnel-top discovery inside coding tools. The task and balance tools are designed for authenticated workflows.
Examples
Ask your assistant natural-language questions. The assistant should use the tools to discover current model slugs and pricing instead of relying on memorized names.
Browse The Catalog
What RunAPI image models are available?
Expected behavior:
- The assistant calls
list_modelswithmodality: "image". - It summarizes the returned model slugs, services, actions, and required fields.
- It avoids quoting stale prices unless it calls
check_pricing.
Search Prompt Examples
Find image prompt examples for a logo.
Expected behavior:
- The assistant calls
search_promptswithmodality: "image"andq: "logo". - It summarizes returned titles, prompt text, model slugs, categories, tags, and example image URLs.
- It uses the selected prompt with
get_model_infobefore creating a task.
Inspect A Model
Show me the required parameters for this model slug: <model-slug>
Expected behavior:
- The assistant calls
get_model_info. - If the response is ambiguous, it chooses the relevant service/action from the returned matches and calls
get_model_infoagain withserviceandaction. - It shows required fields, enum constraints, range constraints, conditional input rules, supported action, and current runtime pricing if available.
- It tells you to choose another slug with
list_modelsif the slug is not found.
Create A Media Task
Generate a square product image with RunAPI. Pick a suitable image model.
Expected behavior:
- The assistant calls
list_modelsto choose a compatible image model. - It calls
get_model_infowith the selected service/action/model to read the parameter descriptions. - It asks for confirmation if the request is expensive, long-running, or a batch.
- It generates one opaque
idempotency_keyfor this logical task and callscreate_task. - It returns task ID, status, output URLs, and cost fields when available.
Wait For Completion
Hosted MCP keeps an asynchronous create_task call with wait: true open as request-scoped SSE for up to 300 seconds. It polls about every 5 seconds, sends an SSE heartbeat about every 15 seconds, and sends standard notifications/progress only when the MCP client supplied a progressToken. The terminal JSON-RPC response arrives in that same stream and then closes it.
The terminal tool result includes matching structuredContent and text content with task_id, final status, completed: true, and the RunAPI result. See examples/create-task-and-wait.arguments.json.
Submit Without Waiting
Create the task but do not wait for completion.
Expected behavior:
- The assistant generates one opaque
idempotency_keyand callscreate_taskwithwait: false. - It returns the task ID.
- You can later ask for status with
get_task.
Replay A Task Creation Safely
create_task requires an opaque idempotency_key from 1 to 512 characters. Generate one new key for each logical task and retain it with the exact service, action, model, and params used for that task.
If the create result is unknown because the connection closed or timed out, do not automatically create another task. Retry only when intended, using the same key with the same input. Reusing the key with different input returns a conflict. Do not derive the key from a JSON-RPC request ID or X-Client-Request-Id.
When Completion Wait reaches its deadline, task creation remains successful. The non-error result has matching structured and text content with task_id, latest status, completed: false, wait_deadline_reached: true, and next_action: "get_task". Continue with get_task; do not create a replacement task.
If polling fails or the connection closes after task creation, the task may still be processing. Continue with get_task; do not create a replacement task.
See examples/create-task.arguments.json for submit-only arguments.
Check Account Balance
Check my RunAPI balance.
Expected behavior:
- The assistant calls
check_balance. - If no key is configured, it calls
loginfor browser login, or explainsRUNAPI_API_KEYfor headless hosts.
Catalog Coverage
The MCP Embedded Catalog is generated from RunAPI's contract snapshot. It includes only models executable by this server release.
| Modality | What To Use |
|---|---|
| Image | list_models with modality: "image" |
| Video | list_models with modality: "video" |
| Audio and music | list_models with modality: "audio" |
| Utility | list_models with modality: "utility" |
Catalog contents can change between releases.
Use list_models for current service/action/model slugs and get_model_info for each model's current constraints.
Pricing
RunAPI pricing is exposed through the check_pricing tool and the public pricing page.
Do not rely on examples in README files for exact prices.
Useful flows:
- Call
list_modelsto find a candidate model. - Call
check_pricingwithservice,action, andmodel. - Show the returned current price or link to runapi.ai/pricing.
Free catalog tools do not create tasks and do not consume account balance.
Platform Setup
Claude Code, Cursor, Windsurf, And VS Code
Run:
claude mcp add runapi -s user -- npx -y @runapi.ai/mcp
Use -s user for a global install available in all projects.
Use -s project when you want Claude Code to write .mcp.json in the repo for team-shared config.
Restart or reload your MCP host after changing MCP configuration.
Compatibility Fallback: Generated Config
Use init only when a host needs a platform-specific JSON file or cannot use the Claude Code MCP command.
Claude Code fallback:
npx @runapi.ai/mcp init claude
This writes .mcp.json in the current directory.
Cursor fallback:
npx @runapi.ai/mcp init cursor
This writes .cursor/mcp.json.
Open Cursor settings to verify the MCP server is enabled.
VS Code fallback:
npx @runapi.ai/mcp init vscode
This writes .vscode/mcp.json.
VS Code uses a top-level servers key and type: "stdio" in generated config.
Windsurf fallback:
npx @runapi.ai/mcp init windsurf
This writes the generated config for the Windsurf target used by the init command.
Roo Code
Run:
npx @runapi.ai/mcp init roo
This writes .roo/mcp.json.
Manual Configuration
Use the example files in examples/ as starting points.
Each platform has slightly different wrapper keys and file paths, but all run the same command:
npx -y @runapi.ai/mcp
Configuration
The server reads configuration in this order:
RUNAPI_API_KEYenvironment variable, useful for headless and CI hosts~/.config/runapi/config.json, created by the MCPlogintool orrunapi login- No key, which still allows free catalog tools
The config file is normally managed by login. A pre-provisioned headless config can use:
{
"apiKey": "your_runapi_key"
}
You can also set a custom base URL for local testing:
{
"apiKey": "your_runapi_key",
"baseUrl": "https://runapi.ai"
}
Do not commit real API keys.
Data Sync
This package ships build-time contract data:
data/contract.json: catalog, actions, model slugs, and input constraints
Refresh data from the RunAPI source tree before a release:
npm run sync:data
Contract updates require a new package release. Pricing is queried from the RunAPI runtime API and does not use a package snapshot.
Development
npm install
npm run typecheck
npm test
npm pack --dry-run
Run the server locally:
npm run dev
Manual initialize smoke test:
printf '%s\n' '{"jsonrpc":"2.0","method":"initialize","params":{"protocolVersion":"2025-03-26","capabilities":{},"clientInfo":{"name":"test","version":"0.1.0"}},"id":1}' | npx tsx src/index.ts
Package Contents
The npm package includes:
- compiled
dist/files - embedded
data/files - platform examples
- eval scenarios, when generated by this repo
- README, changelog, license, and package metadata
It does not include node_modules, .env, local config files, or API keys.
Also Available Via CLI
RunAPI also has a separate command-line client for terminal workflows. Use this MCP server when you want RunAPI available inside an MCP host. Use the CLI when you want direct shell commands, scripts, or CI integration.
License
Licensed under the Apache License, Version 2.0.
RunAPI MCP Fragen
Wie füge ich RunAPI MCP zu Claude Code hinzu?
Führe claude mcp add --env 'RUNAPI_API_KEY=<YOUR_RUNAPI_API_KEY>' mcp -- npx -y @runapi.ai/mcp in deinem Terminal aus. Füge --scope user hinzu, damit der Server in jedem Projekt verfügbar ist.
Wie füge ich RunAPI MCP zu Cursor hinzu?
Öffne ~/.cursor/mcp.json und füge den Server unter mcpServers hinzu, mit der Cursor-Config aus dem Installationsabschnitt oben.
Welche Tools bietet RunAPI MCP?
Er stellt 9 Tools bereit, darunter list_models, get_model_info, list_actions, check_pricing, search_prompts.
Benötige ich einen API-Key, um Modelle zu durchsuchen?
Nein, Discovery-Tools wie das Durchsuchen des Katalogs und das Prüfen der Preise funktionieren ohne API-Key über einen eingebetteten Build-Time-Katalog.
Wie kann ich RunAPI in einer Headless- oder CI-Umgebung nutzen?
Du kannst die Umgebungsvariable RUNAPI_API_KEY setzen, bevor du den MCP host startest.
Kann ich Aufgaben erstellen, ohne auf deren Abschluss zu warten?
Ja, du kannst das Tool create_task aufrufen und den Parameter wait auf false setzen, um sofort eine Task-ID zu erhalten.
Läuft dieser Server die KI-Modelle lokal?
Nein, das Package ist ein reiner Client und betreibt kein lokales Generation-Backend.