This product was not featured by Product Hunt yet. It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).
Product upvotes vs the next 3
Waiting for data. Loading
Product comments vs the next 3
Waiting for data. Loading
Product upvote speed vs the next 3
Waiting for data. Loading
Product upvotes and comments
Waiting for data. Loading
Product vs the next 3
Loading
mcpeye
PostHog for MCP servers. See why your agents fail.
PostHog for MCP servers. Your users are AI agents, and when one can't do something it doesn't file a bug. It quietly gives up or hallucinates a workaround. mcpeye answers the product question: are agents succeeding, and what did they need that you didn't ship? Intent Gap Report ranks the asks your tools couldn't satisfy, plus capabilities no tool covers yet. Session Replay steps through any agent run as one transcript. Goal success rates show what percent succeed. One line, docker compose up.
Hi PH, maker here.
We kept shipping MCP tools and had no real idea if they worked. The thing about building for agents is your users never complain. A human who hits a wall files a bug or churns loudly. An agent just gives up, hallucinates a workaround, and moves on. The failure is real but silent, so you never hear about it. Uptime and tracing tools told us the server was healthy and fast. They couldn't tell us whether agents were getting what they came for.
So we built mcpeye. You add one line to your MCP server (track(server, "your-project-id") in TypeScript, Python, or Ruby) and it captures what agents were trying to do, the args, the result, and the errors. The main feature is the Intent Gap Report: a ranked, deduped list of the asks your tools were called for but couldn't satisfy, plus the capabilities agents wanted that no tool exists for yet (inferred from intent, or captured explicitly when an agent calls the mcpeye_request_capability tool we inject). Every row links to the real transcripts, so it reads like a build-next list your own agents wrote for you. There's also session replay, per-goal success rates, error patterns, and a weekly markdown/CSV report.
It's self-hosted. docker compose up brings up Postgres, Redis, the ingest API, the worker, and the dashboard on localhost:3000 in about a minute. Your prompts and tool payloads stay on your machine. There's no per-call LLM (one only runs later in a worker to cluster sessions, and you bring your own OpenAI or Anthropic key), so overhead is near zero, and you don't need a key to start.
One honest limit: mcpeye sees asks that actually reached your server, not the truly silent client-side misses where an agent never called a tool at all. Better to say that up front.
It's MIT and about two weeks old, so it's young and rough in places. We're committing to turning around bug fixes and feature requests within 24 hours. If you run an MCP server, please try it and tell us where it falls short. Repo: https://github.com/mcpeye/mcpeye
About mcpeye on Product Hunt
“PostHog for MCP servers. See why your agents fail.”
mcpeye was submitted on Product Hunt and earned 3 upvotes and 1 comments, placing #108 on the daily leaderboard. PostHog for MCP servers. Your users are AI agents, and when one can't do something it doesn't file a bug. It quietly gives up or hallucinates a workaround. mcpeye answers the product question: are agents succeeding, and what did they need that you didn't ship? Intent Gap Report ranks the asks your tools couldn't satisfy, plus capabilities no tool covers yet. Session Replay steps through any agent run as one transcript. Goal success rates show what percent succeed. One line, docker compose up.
On the analytics side, mcpeye competes within Analytics and Developer Tools — topics that collectively have 687.5k followers on Product Hunt. The dashboard above tracks how mcpeye performed against the three products that launched closest to it on the same day.
Who hunted mcpeye?
mcpeye was hunted by Mostafa Ali. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
For a complete overview of mcpeye including community comment highlights and product details, visit the product overview.
Mostafa Ali