Kyma API MCP Server

Kyma API is an LLM API gateway that routes one key to 100+ models and publishes measured uptime per model.
Every model behind one endpoint, with public rankings and a spend cap you set.
https://mcp.kymaapi.com/mcp

Install · Tools · Security and spend · Examples · Troubleshooting · Registries

Kyma API is an LLM gateway: one OpenAI-compatible endpoint (plus an Anthropic-compatible one) in front of open and frontier models, with uptime measured per model from real traffic and scheduled probes. This server gives any MCP client the live catalog, prices, rankings, uptime and your own spend as read tools, and one chat tool that is guarded by a per-connection spend cap.

• Hosted, remote, OAuth sign-in. Streamable HTTP with OAuth 2.1, PKCE and dynamic client registration. No API key is pasted into the client.
• A dedicated key per connection. Separate from your REST API keys, capped at $10 per 30 days by default (adjustable from $1 to $500 on the approval screen or at kymaapi.com/integrations), revocable at any time.
• Read tools never charge. Catalog, pricing, rankings, uptime, credits, spend, your usage and your requests are free to call and auto-approved by most clients.
• send_message is priced before it spends. The client must allow it first; max_cost_usd is enforced by deriving a max_tokens that fits the ceiling from the catalog price, so a call above it is refused before anything is sent. Every reply reports what the call cost, what has been spent and the cap.
• Measured uptime, not a promise. get_model_uptime returns the 30-day rate per model from kymaapi.com/status. If a route fails mid-request, Kyma retries the same model on another route.
• A read-only endpoint for reviewers. https://mcp.kymaapi.com/mcp/readonly accepts the same tokens but keeps only read scopes for the request: no send_message, no set_low_balance_alert, whatever the grant was approved with. Metadata: /.well-known/oauth-protected-resource/mcp/readonly.
• Beyond chat. The same Kyma key covers speech-to-text, text-to-speech, embeddings, rerank, image and video through the REST API; list_models shows all of them.

Install

Claude Code

claude mcp add --transport http kyma https://mcp.kymaapi.com/mcp

Run /mcp to sign in. This repository is also a Claude Code plugin (.claude-plugin/plugin.json + .mcp.json).

Claude.ai and Claude Desktop

Settings → Connectors → Add custom connector → paste https://mcp.kymaapi.com/mcp → sign in with Kyma.

Cursor

One click: Add to Cursor

Or in .cursor/mcp.json:

{ "mcpServers": { "kyma": { "url": "https://mcp.kymaapi.com/mcp" } } }

VS Code (GitHub Copilot)

One click: Install in VS Code

Or in .vscode/mcp.json:

{ "servers": { "kyma": { "type": "http", "url": "https://mcp.kymaapi.com/mcp" } } }

Codex CLI

# ~/.codex/config.toml
[mcp_servers.kyma]
url = "https://mcp.kymaapi.com/mcp"

codex mcp login kyma

Cline

Cline → MCP Servers → Configure MCP Servers:

{ "mcpServers": { "kyma": { "url": "https://mcp.kymaapi.com/mcp", "type": "streamableHttp" } } }

Cline CLI: cline mcp add kyma https://mcp.kymaapi.com/mcp --transport streamableHttp --yes. Cline asks you to sign in with Kyma on first use.

Gemini CLI

gemini extensions install https://github.com/kyma-api/kyma-mcp-plugin

The first tool call opens a browser for OAuth sign-in, handled by Gemini CLI.

Grok

• grok.com: Connectors → New Connector → Custom → paste the URL.
• Grok Build: this repository is a Claude Code format plugin; Grok Build loads it as-is.

ChatGPT (Developer Mode)

Settings → Connectors (Plus, Pro, Business, Enterprise, Edu) → add the URL → sign in. ChatGPT connections never see get_topup_link.

Any stdio-only client

{ "mcpServers": { "kyma": { "command": "npx", "args": ["-y", "@kyma-api/mcp-server"] } } }

The bridge runs mcp-remote against the hosted URL, so the tools are exactly the hosted ones. Set KYMA_MCP_KEY=km-… (a Kyma MCP key from kymaapi.com/integrations, not a REST kyma-… key) to skip the browser. Source: bridge/.

Tools

18 tools. 16 appear on a default connection; list_keys and set_low_balance_alert appear when their optional scopes are granted on the approval screen.

Tool	What it does	Scope	Approval
list_models	Live model catalog with prices, context window and capabilities	models.read	auto, read-only
get_model	One model's details	models.read	auto, read-only
list_pricing	Prices across the catalog	models.read	auto, read-only
list_rankings	Top models and apps from real traffic	models.read	auto, read-only
get_model_uptime	Measured 30-day uptime per model	models.read	auto, read-only
recommend_model	Best model and config for Cline, Cursor, Claude Code and other agents	models.read	auto, read-only
search_docs	Search Kyma docs	models.read	auto, read-only
ping	Health check	models.read	auto, read-only
get_credits	Your balance	usage.read	auto, read-only
get_spend	This connection's cap, spent, remaining and reset date	usage.read	auto, read-only
get_usage	Your requests, tokens and cost by model over a window	usage.read	auto, read-only
get_request	One request by id: model served, tokens, cost, routes tried	usage.read	auto, read-only
estimate_cost	What a call would cost before you make it	models.read	auto, read-only
get_transactions	Recent credit ledger entries	usage.read	auto, read-only
get_topup_link	Balance and the billing page (a link, never a checkout)	usage.read	auto, read-only
list_keys	Your REST API keys by name, masked	keys.read (optional)	auto, read-only
set_low_balance_alert	Balance at which Kyma emails you	billing.alerts (optional)	asks once
send_message	A chat completion through any model; the only tool that spends credit; optional max_tokens and max_cost_usd	chat.completions	asks once; no charge without your Allow

Every send_message reply carries cost, spent_usd and spend_cap_usd, plus structuredContent with the model that served it and the routes tried. get_spend shows what is left before you call it.

Security and spend

• What a connection can do: read the catalog, prices, rankings, uptime, your balance and this connection's spend; send chat completions up to its cap; with optional scopes, list your keys masked and set a low-balance alert.
• What it can never do: create or delete API keys, change billing or payment methods, buy credits, register accounts, or see provider internals. Those scopes do not exist on this server.
• Spend cap: each connection gets a dedicated key with its own cap, $10 per 30 days by default, adjustable from $1 to $500. The cap is enforced on the server; when it is reached send_message returns an error until the window resets or you raise the cap at kymaapi.com/integrations.
• Tokens and keys: access tokens live 7 days, refresh tokens 90 days and rotate on use; connection keys are stored hashed (SHA-256) and can be revoked without touching your REST API keys.
• Prompt injection: model output that reaches your agent is untrusted text. Keep your client's confirmation prompt on for send_message, and do not let the agent paste keys or balances into other tools.
• Data: request logs are kept 90 days for billing reconciliation and abuse handling, then deleted. Kyma does not train on customer data. Privacy: kymaapi.com/privacy.

Skills

Three skills teach an agent how to use the tools well: pick-a-model (compare price, measured uptime and fit), check-spend-and-credits (balance, cap, remaining, reset date), send-a-test-completion (one billed call, only after an explicit Allow). They live in skills/ and install into any agent that reads skills:

npx skills add kyma-api/kyma-mcp-plugin

Examples

Prompts you can paste once connected:

• "Which models under $1 per million output tokens have the best 30-day uptime?"
• "Recommend a model and config for Cline, then show its price."
• "How much of this connection's budget is left, and when does it reset?"
• "Send 'summarize this in one line' to qwen-3.6-plus and tell me what it cost."
• "Show the top apps on Kyma rankings this week."

Troubleshooting

• Sign-in loops or "connector failed": remove the connection and add it again; the client's stored token may belong to a revoked connection. Check kymaapi.com/integrations for the active list.
• send_message says the cap is reached: raise the cap at kymaapi.com/integrations or wait for the 30-day window to reset; get_spend shows the date.
• A tool returns 403 insufficient scope: the connection was approved without that optional scope. Reconnect and tick it on the approval screen.
• Stdio bridge cannot sign in: set KYMA_MCP_KEY to a Kyma MCP key (km-…) created at kymaapi.com/integrations; REST kyma-… keys are rejected on purpose.
• Remote connections drop: some clients need a reconnect after long idle periods; the server itself is stateless.

Registries and marketplaces

• Official MCP Registry: com.kymaapi/kyma (server.json in this repo, published with mcp-publisher, namespace verified by DNS on kymaapi.com)
• Smithery: kyma-api/kyma
• npm stdio bridge: @kyma-api/mcp-server
• Gemini CLI extensions gallery: gemini-extension.json and the repo topic gemini-cli-extension
• Submitted, pending review: Cursor Marketplace, xAI Grok Build marketplace, Docker MCP Catalog, Cline Marketplace, mcp.so

About Kyma API

Kyma is an LLM API gateway: open and frontier models, one endpoint, pay per token, with automatic failover across providers and per-model measured uptime published at kymaapi.com/models. New accounts start with free credit.

• Website: https://kymaapi.com · This server's page: https://kymaapi.com/mcp · Docs: https://docs.kymaapi.com · Status: https://kymaapi.com/status
• Support: hello@kymaapi.com · Abuse: report@kymaapi.com
• Changelog: CHANGELOG.md · License: MIT