Context MCP Server tools select shared graphs by graph_uuid
The Context MCP Server tools for shared Context Graphs now take graph_uuid in place of graph_id. The list_graphs tool and the zep://graphs/directory resource return graph_uuid and use cursor pagination with cursor and page_size. A call that sends limit fails. Update agent prompts and saved tool calls that use the previous arguments.
Total limit for custom instructions in each scope
A user, a graph, or the project-wide default now stores at most five custom instructions. The same limit applies to user summary instructions. Zep rejects a request with a new name above the limit with a 400 error. Zep then writes no instructions from the request. An update to a stored name does not add to the total. Before, repeated requests could store more than five instructions in one scope.
Agent model for BYOM accounts
You can now set an agent model for an account or a project that uses Bring Your Own Model (BYOM). The Graph Assistant uses the agent model. Graph ingestion continues to use the primary and fallback models. Open LLM Providers in the account settings or in the project settings to set the agent model.
Batch ingestion keeps fact invalidations and episode sources
When two episodes of one batch resolve to the same fact, Zep now keeps the invalidation and the source episodes of each copy. Before, the fact could stay valid after a later episode invalidated it. The fact could also disappear when one source episode was canceled, although another episode still supported the fact.
Context MCP sign-in with Google Workspace keeps the same user
A Google Workspace sign-in to the Context MCP Server now keeps the identity on the existing user. Before, in some cases, a sign-in created a new user each time the server checked the identity again.