
Lodd.dev
Web Analytics MCP for Agents
PostHog just announceed PostHog Code, which is AI that auto-adds tracking to your codebase. Separate product, own pricing.
This made me realise I might already have something similar by accident. My coding agent has both my codebase and my analytics in the same session via MCP. If I tell it to care about tracking, it sees what it's building, checks what events already exist, and fills the gaps on its own.
The "tell it to care" part was a few lines in my CLAUDE.md, telling it to make sure to add tracking calls, check bounce rates, etc.
First thing it caught was five components I'd shipped over two days with zero event tracking. Comparison pages, pricing buttons, an interactive demo. All user-facing, none tracked.
I don't know if this holds up. It's been a few days. The instructions might be too vague, or the agent might need something more structured to do this reliably. But the gap between "has access to analytics" and "uses analytics while coding" was a lot smaller than I expected.
1 Like
Comment
I completely re-thought my analytics product. Turns out, the AI email summary was the thing people actually liked. A simple analysis of what was happening, shared where you were. The most useful part of the product was the part that meant you never had to open the product.
With lodd.dev, you get your analytics where you work, able querying them whenever it needed to. Not a dashboard with an AI bolted on, but analytics where the AI is the interface.
It's a complete rework, rebuilt as an API with 37 MCP tools optimised for how agents work. Traffic sources, page performance, date comparisons, geographic breakdowns, custom events, the lot. Your agent connects, queries what it needs, and when a human actually wants to see a chart, the agent generates one on the fly. Designed for what an LLM needs: focused, structured, token-saving output.
There's still a website for signing up and managing API keys, but the core product has no GUI. Your agent is the interface.
A month in now. Free up to 2,500 events/month, €9.99/mo for 100K. Tracking script is 2KB, setup is about two minutes if you're already in Claude or Cursor.
I'm picking a pretty tight niche here, since this obviously only works if you already have an AI agent running. If you want to check your traffic the old-fashioned way, Plausible and Fathom are great. But if you're in Claude Code all day anyway and you'd rather your agent just knew your traffic without you alt-tabbing to anything, that's what this is.
1 Like
Comment
You ship something. Then you wonder: did anyone see it?
Earlier this year I built a GDPR-friendly analytics tool for a prototype I was helping with. One script tag, no cookies, no consent banners. It worked, and then I kept going until it became cheapanalytics.com. The core idea is simple: that moment between "I shipped it" and "Did anyone see it?" needs an answer. Not a dashboard you check daily, not funnels and cohorts, just enough signal to feel progress.
It's privacy by default, which means no cookies, no personal data, GDPR compliant from day one. I added weekly email summaries written in plain English, where an AI looks at your traffic and tells you what mattered. You can ignore the dashboard entirely if you want. It costs €3,99 a month for unlimited sites and 100k page views. The emotional truth here is that moment when you realise people are actually looking at your thing. That validation is what you're selling.
It's built for solo founders and indie builders who want first-layer insight without learning analytics. It should be as simple as possible, while still giving you guidance.
I'd love your feedback on the onboarding flow and whether the weekly email actually helps keep momentum. What do you use for analytics on your side projects?

About
Lodd.dev is the result of building a basic analytics product, realising that the only interesting feature was my AI summary, so rebuilding it from scratch to be agent-first. Now it can be part of your AI workflow!


Comment