LogLens is an LLM observability platform built for developers shipping AI products. Add one line of code and every call your app makes to Anthropic or OpenAI gets logged automatically, full prompt, completion, latency, token count, cost, and errors. No proxy setup, no framework lock-in, works with both TypeScript and Python. Built-in cost alerts email you before your bill explodes. Error tracking shows failed calls in red with the exact error type. Free tier, no credit card required.
I built LogLens because I was shipping an AI app and had no idea what was actually happening - prompts were going in, responses were coming out, but I couldn't see costs creeping up or which prompts were slow. Existing tools were either too complex or tied to specific frameworks.
So I built a lightweight SDK that wraps your existing Anthropic or OpenAI client in one line -no refactoring needed. Everything logs automatically to a dashboard where you can search, filter, and inspect every call.
Built the entire thing in a weekend using Claude. Would love feedback on what features matter most to you!
Report
Maker
Quick start —
npm install loglens-sdk
Wrap your client, get your API key from Settings at llm-watch.vercel.app, and every AI call your app makes shows up in your dashboard automatically. Full docs in the Settings page once you sign up
Report
No reviews yetBe the first to leave a review for Loglens
Quick start —
npm install loglens-sdk
Wrap your client, get your API key from Settings at llm-watch.vercel.app, and every AI call your app makes shows up in your dashboard automatically. Full docs in the Settings page once you sign up