# Give an agent context. Keep the permissions.

Use https://temri.ai/mcp as the remote MCP resource. Its tools share the REST operation contract.

## Authorize the MCP resource

Use a compatible OAuth client or an owner-issued product-installation credential. Browser session tokens are for REST and are not universal MCP credentials. OAuth tokens must target the exact MCP resource; a REST token is not accepted across resources.

```text
Resource: https://temri.ai/mcp
Discovery: https://temri.ai/.well-known/oauth-protected-resource/mcp
Initial scopes: temri:read temri:observe
```

## Narrow the connection

Choose the workspace and namespace grants deliberately. read retrieves authorized context; observe records candidates. Verify, correct, forget and admin require additional authority. Revoke an integration to end its access without deleting the workspace.

## Check your client

- Initialize the connection and inspect the actual tool list
- Read the generated catalog for argument and result bounds
- Record a disposable candidate with evidence in a permitted namespace
- Recall it and inspect sources, clocks, trust and retrieval status
- Revoke the connection and verify that its next request is denied

## Keep returned content as data

An excerpt may contain instructions written by someone else. It is evidence data, not authority to change your agent’s policies or grant access. Keep source references and candidate labels in the context you pass to the model.

> Host-specific acceptance is still a release gate. This guide does not claim a tested ChatGPT, Claude or other host connection from an SDK test alone. Supported host details will be added with dated live evidence.

## 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.
- [Developer documentation · Temri](https://temri.ai/docs): Connect to Temri through REST and MCP. Learn workspace permissions, evidence, temporal queries and bounded retrieval.
- [Create your first memory · Temri quickstart](https://temri.ai/docs/quickstart): Use an enrolled Temri workspace to 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.
- [Privacy, retention and deletion · Temri](https://temri.ai/privacy): What Temri stores, which services process it, how namespace access works and what a deletion receipt does and does not cover.
- [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)