Runs and failures
Every run for every agent, which runs failed and why, and which agents haven't found anything yet.
See how your AI agents are doing
Agentis Platform logs every agent run. See what each agent found, whether the user got notified, what users thought of the results and what it all cost.
What you can track
Your products send their agent runs to the platform. This is what you see in the dashboard.
Findings and feedback
How often each agent finds something, why empty runs came back empty, and the thumbs up or down users gave the results.
Notifications
Check that results were sent to users, and see which notifications were blocked or failed.
AI costs
Spend per model, per pipeline step and per run, plus every search and page fetch a run made.
How it works
Add the SDK to your product and send an event when an agent runs. The results show up in the dashboard.
- Sendtrack()Call track() from your product when an agent runs, with what it found and what it cost.
- Store/v1/eventsEach event is saved under its project, agent and customer.
- ViewdashboardSee runs, findings, notifications and costs for each agent.
Want to use it for your agents?
The platform is in beta. Get in touch and we'll set you up with a project and an API key.