Call Claude models directly — run AI conversations, tool-use chains, batch jobs, and model comparisons from any agent.
Use with AI AssistantsMCP
Connect Claude, Cursor, or any MCP-compatible client — then call tools directly
① Add This MCP Server
Paste into your AI client config — then all its tools are available instantly.
{
"mcpServers": {
"anthropic": {
"url": "https://mcp.aerostack.dev/s/aerostack/mcp-anthropic",
"headers": {
"Authorization": "Bearer YOUR_AEROSTACK_TOKEN"
}
}
}
}Replace YOUR_AEROSTACK_TOKEN with your API token from the dashboard.
② Call a Tool
Ask your AI assistant to call a specific tool, or send raw JSON-RPC:
Natural Language Prompt
“Use the _ping tool to verify anthropic credentials by calling a lightweight read endpoint. used internally by aerostack to validate credentials”
Using a Workspace?
Add this MCP to your Workspace — your team shares one token, secrets are stored securely, and every AI agent in the workspace can call it without per-user setup.
Give your AI agents direct access to Claude — send messages, run batches, manage models, and administer your Anthropic organization.
Anthropic's API powers some of the world's most capable AI models. This MCP server exposes the full Anthropic API surface to your agents: from sending individual messages and tool-calling conversations to processing thousands of requests in efficient batches and administering your organization's workspaces and API keys.
Live endpoint: https://mcp.aerostack.dev/s/aerostack/mcp-anthropic
| Tool | Description |
|---|---|
| create_message | Send a message to a Claude model and get a response with full parameter control |
| create_message_with_tools | Send a message with tool definitions so Claude can respond with tool use blocks |
| count_tokens | Count input tokens for a request without sending it — use to estimate cost |
| create_message_batch | Submit a batch of message requests for async processing — returns a batch ID |
| list_models | List all available Claude models with IDs and creation dates |
| get_model | Get details for a specific Claude model by ID |
| list_batches | List message batch jobs with status, request counts, and timestamps |
| get_batch | Get status and result URL for a specific message batch |
| cancel_batch | Cancel an in-progress message batch |
| list_workspaces | List all workspaces in your organization (Admin API key required) |
| get_usage | Get API usage and billing data for your organization (Admin API key required) |
| list_api_keys | List all API keys in your organization (Admin API key required) |
| Variable | Required | Description | How to Get |
|---|---|---|---|
| ANTHROPIC_API_KEY | Yes | Anthropic API key (standard or Admin) | console.anthropic.com → API Keys |
ANTHROPIC_API_KEY under Project → SecretsOnce added, every AI agent in your workspace can call Claude models automatically — no per-user setup needed.
"Ask Claude to summarize this document using the claude-opus-4-5 model"
"Count how many tokens this conversation will use before sending it"
"Submit a batch of 500 classification requests and check the status"
curl -X POST https://mcp.aerostack.dev/s/aerostack/mcp-anthropic \
-H 'Content-Type: application/json' \
-H 'X-Mcp-Secret-ANTHROPIC-API-KEY: your-key' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"create_message","arguments":{"messages":[{"role":"user","content":"Hello!"}]}}}'
MIT
Live Endpoint
https://mcp.aerostack.dev/s/aerostack/mcp-anthropic
Sub-50ms globally · Zero cold start
@aerostack
Pre-built functions for the most common MCP tool patterns. Clone, extend, and deploy.
by @aerostack
Design creation, AI generation, editing, export, assets, brand kits, folders via Canva's official MCP
by @aerostack
Conversations, contacts, tickets, articles via Intercom's official MCP
by @aerostack
Deployments, projects, domains, env vars, logs via Vercel's official MCP
by @aerostack
Read, send, and search Gmail messages — manage labels, drafts, and attachments with full inbox control.
by @aerostack
Manage audiences, campaigns, and subscriber tags in Mailchimp — automate your email marketing with AI.
by @aerostack
Run chat completions, generate embeddings, create images, and moderate content via the OpenAI API.
Yes. The create_message tool accepts a model ID, system prompt, and messages array. An orchestrator agent can spawn a Haiku sub-call for cheap classification and a Sonnet call for synthesis — all within one workflow, billed to your Anthropic key.
Use create_message_batch to submit, then list_batches or get_batch to check status. When the batch reaches succeeded, retrieve individual results via get_batch — Claude can automate the poll loop without manual API calls.
Yes. get_usage returns aggregated token counts and cost data per workspace. count_tokens lets you pre-flight a prompt before sending it, so Claude can warn you if a context window is about to exceed a budget threshold.
Yes. list_models returns all available model IDs and their context window sizes. Claude can dynamically select the cheapest model that fits the token budget rather than relying on a hardcoded string like claude-3-haiku-20240307.