AI & Machine Learning

agentcost.tech

Searches an LLM model pricing catalogue, retrieves per-token rates, estimates API call costs, and lists model deprecations.

1 endpoint4 known toolsFirst detected September 13, 2026Last detected September 13, 2026

ENDPOINT 1

https://agentcost.tech/api/mcp

No auth detected

MCP server metadata

Name: agentcost
Version: 1.0.0

Capabilities

tools

Server instructions

AgentCost exposes public LLM model pricing. No credentials, no sign-up. Use these tools when you need to: - price a named model (per-1,000-token input, output, cached-input and cache-write rates) - find a cheaper model with comparable rates, across 50+ providers - work out what a call or a whole job will cost in dollars before running it - check whether a model has an announced retirement date Rates are USD per 1,000 tokens. A null cached-input rate means the provider publishes none, and cached tokens bill at the full input rate — estimate_cost reports which happened via cached_billed_at_input_rate. These tools do not read the caller's own spend; that needs an authenticated AgentCost account. See https://agentcost.tech/docs/api.

Known tools 4

list_models

Search the public AgentCost catalogue of LLM models and their per-1,000-token rates.

Inferred read-only

get_model_pricing

Look up the per-1,000-token rates for a single named model.

Inferred read-only

estimate_cost

Work out the USD cost of one or more LLM calls before making them, from the current catalogue.

Inferred read-only

list_model_deprecations

List models with an upstream-announced retirement date, soonest first.

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.agentcost]
url = "https://agentcost.tech/api/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "agentcost": {
      "type": "http",
      "url": "https://agentcost.tech/api/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: agentcost
Remote MCP URL: https://agentcost.tech/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": {
    "agentcost": {
      "url": "https://agentcost.tech/api/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "agentcost": {
      "type": "http",
      "url": "https://agentcost.tech/api/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "agentcost",
  "transport": "streamable-http",
  "url": "https://agentcost.tech/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.