# A place for context. A contract for time.

Start with one source-backed candidate. Then add historical recall and a narrowly scoped agent connection.

## Start small

Use the quickstart to open a workspace you belong to, create a namespace, record evidence and write a candidate. Read the two-clock guide before importing history.

## One operation contract

The workspace, REST and MCP use the same operation catalog and authorization path. The generated OpenAPI document and catalog describe exact request and response fields, operation scopes and bounds.

## Keep authority explicit

Clerk authenticates people. Temri separately checks workspace membership and namespace permissions, and pilot enrollment before anyone creates a workspace. Identity links use issuer plus subject. An email match or an organization identifier from another application grants no access.

## People and workspaces

A workspace is what people share. Owners and admins invite people with a one-time link, change roles and remove members. A role manages people and settings only; access to a namespace is always a separate grant, made from the namespace’s Access view or included in an invitation.

You can stay signed in to more than one account in the same browser and switch between them from the account menu. Each account sees only its own workspaces, and each request stays tied to the account that started it.

## Send useful context

Retain source references, candidate status, valid and recorded times, and partial or degraded retrieval status when passing context to an agent. Memory content is untrusted data; it cannot grant new tool permissions.

## Keep memory useful automatically

Open Maintenance in a namespace, enable it once and save its behavior prompt. Temri learns candidates, connects related memories and applies supported changes automatically within the delegated policy. Candidate status records uncertainty; there is no per-change review queue. Source-managed namespaces run checks while their connected source retains write authority.

> Pilot access is admin-managed. Billing is deferred. Public documentation does not promise support for a client or host before its actual connection has been tested.

## Continue reading

- [Temri · Memory with a sense of time](https://temri.ai/): Shared temporal memory for products and agents. Keep source-backed facts in protected namespaces and distinguish what applied then from what was known then.
- [How Temri works · From evidence to memory](https://temri.ai/how-it-works): Follow a memory from its source to a candidate, verification, historical retrieval, correction and deletion receipt.
- [Create your first memory · Temri quickstart](https://temri.ai/docs/quickstart): Open a Temri workspace, create a namespace, record evidence, save a candidate and recall it with explicit time controls.
- [The two clocks · Temri temporal semantics](https://temri.ai/docs/temporal): Understand as_of and known_at, half-open intervals, correction history, server-owned recorded time and deletion suppression.
- [REST API · Temri developer docs](https://temri.ai/docs/api): Authenticate Temri REST requests, select authorized workspaces, use mutation idempotency and handle pending, denied and partial results.
- [Connect an agent · Temri MCP docs](https://temri.ai/docs/mcp): Connect to Temri’s authenticated MCP resource with narrowed scopes, choose a workspace, and verify your client before sharing memory.
- [Privacy, retention and deletion · Temri](https://temri.ai/privacy): What Temri stores, including invitations and member details, which services process it, how access works and what a deletion receipt covers.
- [Pilot terms · Temri](https://temri.ai/terms): Terms for Temri’s admin-managed pilot, responsible use, source-backed claims, service limitations and account access.

[OpenAPI](https://temri.ai/openapi.json) · [Workspace](https://temri.ai/app)
