Nothing on this page is aspirational. Anything not yet built is marked as such and sits below the line.
Everything it does today.
Six deep-dives, and the platform underneath them. This page lists what is built and serving traffic, the roadmap tier is at the bottom, kept separate on purpose.
Start with one of these.
The first two are the argument: a dashboard the person who paid for the agent is allowed to open, and a ticket loop that turns their complaint into something you can reproduce. The rest is how you hand it over.
Analytics
5 KPI cards, 11 charts.
Drag-and-drop, filtered by date range and environment, free, with no card required.
Learn moreTickets
Now it's a ticket, tied to the exchange.
Anchored to the conversation, the feedback, or the query it's about, with a priority, tags and a threaded discussion.
Learn moreTranscripts
Every message, in order.
Tool calls, durations and errors alongside what the human actually typed.
Learn moreWhite-label
Their name on it.
Your client's subdomain, logo and colours, in light and dark.
Learn moreEmbed
Inside your product.
A scoped token and an isolated origin, personal data off by default.
Learn moreReports
PDF, from a snapshot.
A named report with its date range frozen, exported from a per-agent template.
Learn moreReal and shipped, these simply do not carry a page of their own yet.
What comes with all of them.
| Capability | What it is |
|---|---|
| Message templates | A per-agent HTML block bound to message metadata, so an order lookup renders as an order card in the transcript instead of a JSON blob. Sanitised server-side. |
| Feedback | Sentiment and category, scoped to a conversation, an agent or the product, collected from your widget or the API, and promotable into a ticket. |
| Environments | Production and Development on every agent. Reports cover production only, so your test runs never reach a client. |
| Team & roles | Owner, Editor and Viewer with route-level enforcement, invitations, and ownership transfer that keeps a record. |
| API keys | Scoped read-only or full access, per agent, revocable, with expiry and last-used tracking. |
| Span archive & REST API | Conversations, messages, action logs and usage rows are projections; the raw spans are kept and readable. |
| Chat migration | Import existing history from a JSON export, so a client's dashboard doesn't start at zero. |
| Light & dark | Throughout, including the white-label palette. |
Or start from the problem instead.
The same features, read from one angle at a time.
And one thing that is still a plan.
AI Agent Analytics, our own agents reviewing yours. None of it is implemented, so it is kept off the grid above and given its own bento here, deliberately smaller.
Automated insight digests
A standing summary of what changed.
Anomaly detection
Flagged before a client notices.
Conversation quality scoring
The ones worth reading rise to the top.
Suggested tickets
Drafted from a bad exchange, ready to raise.
The whole dashboard, free.
- Full dashboard on the free plan
- No card required
- Model spans never carry prompts