{"name":"tickerr","version":"1.1.0","description":"Real-time LLM model pricing (300+ models), API status, and inference performance for agent routing. Get input/output token costs updated twice daily via OpenRouter API, live operational status checked every 5 minutes, and p50/p95 latency from authenticated API calls. Use compare_pricing + get_model_performance together to route to the cheapest available model within your latency budget. Includes agent-reported failure signals - agents contributing to Tickerr receive live routing recommendations: how many other agents are seeing the same issue and which model to fall back to.","transport":"streamable-http","install":{"claude_code":"claude mcp add tickerr --transport http --url https://tickerr.ai/mcp","cursor":"{ \"mcpServers\": { \"tickerr\": { \"url\": \"https://tickerr.ai/mcp\" } } }","npm":"npx -y tickerr-mcp"},"docs":"https://tickerr.ai/mcp-server"}