Integration

Charm Hyper integration

Charm Hyper is Charm's inference API for coding models (DeepSeek, Kimi, GLM, MiniMax, Qwen and others, plus the Prism routing model), speaking both the OpenAI and the Anthropic formats. List the models with their context window, vision support, reasoning levels and per-million-token prices, send chat completions (with function calling), OpenAI Responses requests or Anthropic Messages requests, and read the team's Hypercredit balance. Requests are paid in Hypercredits (1 Hypercredit is 5 US cents of inference): subscription credits refresh weekly, prepaid credits are spent after them, and HTTP 402 means the balance is insufficient. Streaming is not exposed. Authentication is a Hyper API key (it starts with sk-hyper-) from the Hyper dashboard, sent as a Bearer token in the Authorization header.

Connect Charm Hyper How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Charm Hyper.Built from the 5 Charm Hyper tools below, on the connection you add once.

What you can do with Charm Hyper

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Each badge shows the default agent permission of an endpoint (allowed: an agent calls it without asking; ask first: you are asked before an agent calls it, when the agent or chat asks permission for sensitive actions; never: no agent may call it), which every user can change endpoint by endpoint in their credential settings.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Chat completion, 1 operation

create_chat_completionCreate chat completionPOSTAgents: ask firstSends messages to a Hyper model in the OpenAI Chat Completions format and returns the reply in choices[0].message, with token usage and the cost in Hypercredits. Spends Hypercredits.

Credit balance, 1 operation

get_credit_balanceGet credit balanceGETAgents: ask firstReturns the current Hypercredit balance of the team that owns the API key (1 Hypercredit is 5 US cents of inference).

Message, 1 operation

create_messageCreate messagePOSTAgents: ask firstSends messages to a Hyper model in the Anthropic Messages format and returns the content blocks of the reply with stop_reason and token usage. Spends Hypercredits.

Model, 1 operation

list_modelsList modelsGETAgents: ask firstLists every Hyper model with its context window, maximum output, vision support, reasoning effort levels and per-million-token prices in US dollars. Pass a returned id as model to the chat, response or message tools.

Response, 1 operation

create_responseCreate responsePOSTAgents: ask firstSends a prompt or a list of messages to a Hyper model in the OpenAI Responses format and returns the generated output items. Spends Hypercredits.

Use Charm Hyper from Claude Code, Cursor or Codex

Connect Charm Hyper in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Charm Hyper operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Charm Hyper

Create your workspace
Sign up free. Nothing to install.
Connect Charm Hyper once
Add your Charm Hyper connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.