AI & Machine Learning

heavymill.ai

MCP server for AI advisor model catalog and cost/budget management, offering model listings, provider health checks, pricing tables, and cost estimates.

1 endpoint13 known toolsFirst detected September 11, 2026Last detected September 11, 2026

ENDPOINT 1

https://mcp.heavymill.ai/api/mcp

No auth detected

MCP server metadata

Name: heavy
Version: 0.1.0

Capabilities

tools

Known tools 13

list_advisor_models

Return the advisor model catalog for consult_advisor — tiers, display names, labs, panel defaults, pricing hints, product facts, and suggested_presentation guidance.

Inferred read-only

check_advisor_catalog_health

Free health check: verify consult_advisor catalog model IDs resolve on OpenRouter and Cerebras (listed on provider /models).

Inferred read-only

get_pricing

Return the full pricing table for all Heavy Mill skills as structured JSON.

Inferred read-only

estimate_cost

Estimate the exact cost of a Heavy Mill skill call before running it.

Inferred read-only

budget_check

Check whether a proposed multi-step workflow fits within a budget ceiling before running any skills.

Inferred read-only

get_quickstart

Return a structured quickstart guide for using Heavy Mill.

Inferred read-only

get_job_status

Check or wait for the result of an async job started with webhook_url.

Inferred read-only

check_wallet_status

Check wallet balance and optionally confirm you can afford a specific tool call — in one request.

Inferred read-only

create_api_key

Create a revocable Heavy Mill API key (hm_sk_...) bound to the funded wallet on this session.

Potential side effects

revoke_api_key

Revoke a Heavy Mill API key by key_id so it can no longer spend.

Inferred read-only

consult_advisor

Paid multi-model advisor panel.

Inferred read-only

fast_infer

Cerebras-accelerated inference at $0.01–$0.02/call without consuming your context window.

Inferred read-only

fast_infer_batch

Batch Cerebras inference: up to 50 prompts in one round-trip.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.heavy]
url = "https://mcp.heavymill.ai/api/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "heavy": {
      "type": "http",
      "url": "https://mcp.heavymill.ai/api/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: heavy
Remote MCP URL: https://mcp.heavymill.ai/api/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "heavy": {
      "url": "https://mcp.heavymill.ai/api/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "heavy": {
      "type": "http",
      "url": "https://mcp.heavymill.ai/api/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "heavy",
  "transport": "streamable-http",
  "url": "https://mcp.heavymill.ai/api/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Loading Trust v2 evidence…

Checking the associated registrable domain. The BuiltWith key remains server-side.

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.