Driftline
Checks OpenAI, Anthropic and Google model IDs for retirement dates before agents use them
- Pricing
- Freemium
- Platforms
- CLI, API
- Runs
- Locally
- Version
- 0.11.0
- Listing updated
- 3 days ago
About Driftline
Driftline is a freemium MCP server for ai models and llm tooling. It runs locally on your machine and is started by your AI client.
Coding agents often write model IDs from training data, and many of those models are already retired or about to be. Driftline scans a repository and reports each ID with status (retired, at-risk, deprecated, active or unknown), retirement date, replacement and whether the provider silently routes calls to another model after that date. Exit code 1 means a retired model is in use.
The tracked list covers 257 model IDs and was checked on 13 November 2026, with daily refreshes from provider pages. As an MCP server it works offline from a bundled copy when the network is down, and needs no key or account. The one-command form is npx -y @aidriftline/driftline . --json.
The site suggests pasting a rule into AGENTS.md or CLAUDE.md so agents check before adding a model. A GitHub Action adds PR checks and warnings before each retirement, with prompt drift tests on a Pro plan.
Install
Run in a terminal
claude mcp add driftline -- npx -y @aidriftline/driftlineAdd to claude_desktop_config.json
{
"mcpServers": {
"driftline": {
"command": "npx",
"args": [
"-y",
"@aidriftline/driftline"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"driftline": {
"command": "npx",
"args": [
"-y",
"@aidriftline/driftline"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"driftline": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@aidriftline/driftline"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"driftline": {
"command": "npx",
"args": [
"-y",
"@aidriftline/driftline"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"driftline": {
"command": "npx",
"args": [
"-y",
"@aidriftline/driftline"
]
}
}
}Tools this server provides (3)
Taken from the project documentation; the running server may differ.
- check_models
- Returns status, retirement date and replacement for a list of model IDs.
- scan_project
- Finds retired and retiring model IDs in a folder, with file and line.
- upcoming_retirements
- Lists what retires or gets re-routed in the next N days.
What it does
- Retirement status
- Reports retired, at-risk, deprecated, active or unknown per ID.
- Replacement and routing
- Names the replacement and flags silent routing to another model.
- Project scan
- Lists IDs in a folder with file and line.
- No key needed
- Runs without an account or API key.
- Offline fallback
- Uses a bundled list when offline.
- CI option
- A GitHub Action checks every pull request.
Good for
- Stop an agent from introducing a retired model
- Find models in a repo that retire soon
- Plan migrations ahead of provider deadlines
- Fail a pull request that uses a retired ID
Pricing
Freemium
MCP server and scan free; Pro adds prompt drift tests, price not published
Prices change. Check aidriftline.com before you buy.
Driftline FAQ
Does it need an API key?
No, the scan and MCP server need no key or account.
What does routes: true mean?
The provider forwards requests for that ID to another model after the date.
How often is data refreshed?
Daily against provider pages.
Is the GitHub Action free?
The scan is free; drift tests on your prompts need Pro.
Is Driftline free?
Driftline has a free plan, with paid plans for more usage or features.
How do I add Driftline to Claude Code?
Run this command in your terminal: claude mcp add driftline -- npx -y @aidriftline/driftline
What are the best alternatives to Driftline?
The closest alternatives to Driftline are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Driftline

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools