MCP tracing + send_feedback instantly revealed how agents really use an MCP server
indragie · x · 2026-11-07
indragie reports that adding MCP tracing, a sendfeedback tool, and Patterns (via Braintrust) to their MCP server immediately surfaced how people use it, what's missing, and what's broken.
Key points from Braintrust's post:
- Tracing records inputs, results, timing, and errors per MCP call; supports official MCP SDKs and FastMCP (JS/Python)
- A "successful" tool call can still return wrong data — traces let you compare request vs response and catch silent bugs like ignored environment filters
- telemetry.intent lets agents explain why they're calling; sendfeedback lets them report problems
- Patterns cluster recurring behaviors; Loop supports open-ended investigation
A directly reusable observability setup for anyone shipping an MCP server.
More from coding & agent
- Vercel Tracing goes GA on all plans, enabled by default in Next.js 16.5 — cramforce · 2026-11-07
- Fine-tuned Qwen3.5-4B router model trained on 17.6k rows for about $5.50 on one A100 — MaziyarPanahi · 2026-11-07
- Access your coding agent's localhost via Tailscale instead of waiting for Vercel deploys — chongdashu · 2026-11-07
- The Agentic Web: personal AI assistants that understand what every website can do — JnBrymn · 2026-11-07
- GPT-Image-2.6 lands on Netlify AI Gateway; demo turns any birth city into a looping poster — thisiskp_ · 2026-11-07
- Lattice: open-source local LaTeX workspace with bring-your-own AI agent — LINJIEFUN · 2026-11-07