# Trust Engine AI: model retirements and agent cost

> Trust Engine AI: model retirements and agent cost is a free MCP server for ai models and llm tooling. Free MCP server for AI model retirement dates, repo model scans and usage-cost checks. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Free
- **Platforms:** CLI, Linux, macOS, Windows
- **Website:** https://agenttrustapi.com
- **Repository:** https://github.com/dinevdecor/trust-engine-mcp
- **MCP Registry name:** io.github.DinevDecor/trust-engine-mcp
- **Categories:** AI models and LLM tooling, developer tools
- **Listing updated:** 2026-11-14

## About

The first job is catching models that are about to disappear. A scan tool walks the working directory, finds model ids that appear in a retirement calendar with file and line, and checks them. Retirement dates cover Anthropic, OpenAI, Google Gemini API and Vertex AI, AWS Bedrock and Mistral, each with an official source URL and the date it was checked.

A second tool reads an OpenAI or Anthropic usage or cost export locally and breaks spend down by model and cost driver, with fixes you can make yourself. The README also exposes numbers from a public benchmark of one coding agent on 27 exercises and savings estimates only for the model pairs it measured.

The server is read-only, needs no key or account, sends only GET requests to agenttrustapi.com and skips .env files. The README warns that dates are copied from vendor pages, so follow the source link before acting.

## Features

- **Repo model scan:** Find model ids in code with file and line numbers.
- **Retirement lookup:** Dates per platform, days left and the vendor's replacement.
- **Filtered calendar:** List upcoming deprecations by provider, model or date.
- **Usage cost analysis:** Read a usage export locally and show spend by model.
- **Offline mode:** Set TRUST_ENGINE_OFFLINE=1 to use only the bundled snapshot.
- **CLI and Action:** The same scan runs as a command and a GitHub Action.

## Good for

- Check a repo for deprecated model ids before a release
- Find when a model you depend on retires on each cloud
- Look at where agent spend is going
- Estimate savings for a measured model pair
- Draft a request for a pilot on your own tasks

## Install

### Claude Code (Terminal)

```sh
claude mcp add trust-engine-mcp -- uvx trust-engine-mcp
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "trust-engine-mcp": {
      "command": "uvx",
      "args": [
        "trust-engine-mcp"
      ]
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "trust-engine-mcp": {
      "command": "uvx",
      "args": [
        "trust-engine-mcp"
      ]
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "trust-engine-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "trust-engine-mcp"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "trust-engine-mcp": {
      "command": "uvx",
      "args": [
        "trust-engine-mcp"
      ]
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "trust-engine-mcp": {
      "command": "uvx",
      "args": [
        "trust-engine-mcp"
      ]
    }
  }
}
```

## Tools

- `scan_for_models`: Finds model ids from the calendar in the working directory with file:line.
- `check_my_models`: Retirement date per platform, days left, replacement and source for each model.
- `model_retirements`: Upcoming retirements and deprecations, filtered, with sources.
- `analyze_usage`: Spend by model from a usage or cost export.

## FAQ

### Does it need an API key?

No key or account is needed.

### Does it upload my code?

The scan is read-only and local; the server only sends GET requests to agenttrustapi.com.

### How reliable are the dates?

They are copied from vendor pages, so the README tells you to follow each source link.

### What do I need to run it?

Python 3.10 or newer and uv, or pip install trust-engine-mcp.

### Is Trust Engine AI: model retirements and agent cost free?

Yes. Trust Engine AI: model retirements and agent cost is free to use.

### How do I add Trust Engine AI: model retirements and agent cost to Claude Code?

Run this command in your terminal: claude mcp add trust-engine-mcp -- uvx trust-engine-mcp

### What are the best alternatives to Trust Engine AI: model retirements and agent cost?

The closest alternatives to Trust Engine AI: model retirements and agent cost are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Concord MCP](https://ai-powered.com/mcp/concord-mcp): Concord MCP is a freemium MCP server for developer tools.

---
Source: https://ai-powered.com/mcp/trust-engine-ai-model-retirements-and-agent-cost
