Langfuse + Memtro

Only when you ask, and only what it has been granted: add a comment to a trace, observation, session or prompt; add a score to a trace, observation or session; create a new prompt version (labels such as production change what apps serve); move labels such as production onto a prompt version; create a dataset; add an item to a dataset; delete a dataset item and its run results (permanent); create a score config; create an annotation queue; add a trace, observation or session to an annotation queue; mark an annotation queue item pending or completed; delete a score (permanent); delete a trace (permanent); delete up to 50 traces by id (permanent). Every other tool reads, searches or lists. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.

A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.

Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.

Public key and secret key. In Langfuse open your project → Settings → API Keys → Create new API keys. The pair has every right on its project with no scopes, so make one for Memtro alone. Copy the sk-lf- secret key, shown once.

Yes. Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in Langfuse, choosing the same kinds; the tab shows Langfuse's own steps. A job set to start "When an event arrives" then runs once for each matching event, with the event as its input. The run treats the event as outside data, never as instructions, and anything the run remembers is kept as unverified.