io.github.lazymac2x/ai-rate-limit-tracker is an MCP server published by io.github.lazymac2x, hosted at api.lazy-mac.com. When last checked on 2026-11-21 it was working: it answers and lists its tools. It declares 4 tools, including get_provider_limits, list_providers, track_usage, get_forecast.
Cloudflare Workers MCP server: ai-rate-limit-tracker
160 connections through 5 shared values. Tap any node.
Get published rate limit specs (RPM, TPM, RPD, context window) for an AI provider and optional model.
| Parameter | Type | Description |
|---|---|---|
| model | string | Model name (optional) |
| provider * | string | openai | anthropic | gemini | groq | cohere | mistral | together | deepseek |
{
"name": "get_provider_limits",
"description": "Get published rate limit specs (RPM, TPM, RPD, context window) for an AI provider and optional model.",
"inputSchema": {
"type": "object",
"required": [
"provider"
],
"properties": {
"model": {
"type": "string",
"description": "Model name (optional)"
},
"provider": {
"type": "string",
"description": "openai | anthropic | gemini | groq | cohere | mistral | together | deepseek"
}
}
}
}None yet. Changes appear after a server's second daily check.
Squares are this server's values; dots are other servers sharing them. A shared address or publisher is a lead, not proof of a common operator.
Inferred from the host name and headers: what a server runs on, not who runs it.
List all AI providers supported by the tracker with their models and default limits.
No parameters.
{
"name": "list_providers",
"description": "List all AI providers supported by the tracker with their models and default limits.",
"inputSchema": {
"type": "object",
"required": [],
"properties": {}
}
}Record an AI API call and get real-time usage counters plus warnings if approaching limits.
| Parameter | Type | Description |
|---|---|---|
| model | string | |
| api_key * | string | Your tracker API key |
| provider * | string | |
| latency_ms | integer | |
| tokens_used | integer |
{
"name": "track_usage",
"description": "Record an AI API call and get real-time usage counters plus warnings if approaching limits.",
"inputSchema": {
"type": "object",
"required": [
"api_key",
"provider"
],
"properties": {
"model": {
"type": "string"
},
"api_key": {
"type": "string",
"description": "Your tracker API key"
},
"provider": {
"type": "string"
},
"latency_ms": {
"type": "integer"
},
"tokens_used": {
"type": "integer"
}
}
}
}Forecast when rate limits will be exhausted and get scheduling recommendations.
| Parameter | Type | Description |
|---|---|---|
| model | string | |
| api_key * | string | |
| provider * | string | |
| rate_rpm | number | Your actual request rate (requests/min) |
{
"name": "get_forecast",
"description": "Forecast when rate limits will be exhausted and get scheduling recommendations.",
"inputSchema": {
"type": "object",
"required": [
"api_key",
"provider"
],
"properties": {
"model": {
"type": "string"
},
"api_key": {
"type": "string"
},
"provider": {
"type": "string"
},
"rate_rpm": {
"type": "number",
"description": "Your actual request rate (requests/min)"
}
}
}
}Names, descriptions and schemas were written by the server and are shown as data.
Shared addresses on big platforms (Cloudflare, Vercel and the like) put unrelated servers together; those are marked shared_infrastructure. A match is a pivot, not a link between operators.