Integration

Zep integration

Zep memory service for conversational agents: record the messages of a thread against a user, then read back a ready made context block that carries the facts and entities Zep extracted from every past conversation, rather than replaying the transcript. The same knowledge graph can be fed directly with text or structured data and searched on its own. The key travels in the Authorization header behind an Api-Key prefix.

Connect Zep How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Zep.Built from the 18 Zep tools below, on the connection you add once.

Common Zep use cases

cleaning up test conversations
auditing which conversations exist in a project
creating the memory owner when someone signs up
confirming a user exists before opening a thread
showing a one line recap of a conversation in a queue
building a product knowledge base every agent can search
confirming a shared graph exists before ingesting into it
honouring a request to erase everything held about a person
finding the identifier of a shared graph before searching it
opening a thread when a user starts a new chat with an agent

What you can do with Zep

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Graph, 5 operations

add_graph_dataAdd graph dataPOSTFeeds data into a knowledge graph outside of any conversation.
create_graphCreate graphPOSTCreates a knowledge graph shared across users.
get_graphGet graphGETReads the definition of one standalone knowledge graph.
list_graphsList graphsGETReads the standalone knowledge graphs of the project.
search_graphSearch graphPOSTSearches the knowledge graph of a user or of a standalone graph.

User, 5 operations

create_userCreate userPOSTCreates a user to hold memory across conversations.
delete_userDelete userDELETEDeletes a memory user and everything held about them.
get_userGet userGETReads the profile of one memory user.
list_usersList usersGETReads the memory users of the project.
update_userUpdate userPATCHUpdates the profile of a memory user.

Thread, 4 operations

create_threadCreate threadPOSTCreates a conversation thread for a user.
delete_threadDelete threadDELETEDeletes a conversation thread.
list_threadsList threadsGETReads the conversation threads of the project.
list_user_threadsList user threadsGETReads every conversation thread belonging to one user.

Message, 2 operations

add_messagesAdd messagesPOSTRecords new messages on a conversation thread.
get_messagesGet messagesGETReads the raw messages of a conversation thread.

Context, 1 operation

get_contextGet contextGETReads the memory block to put in front of a model for this thread.

Summary, 1 operation

get_thread_summaryGet thread summaryGETReads the running summary of a conversation thread.

Use Zep from Claude Code, Cursor or Codex

Connect Zep in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Zep operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Zep

Create your workspace
Sign up free. Nothing to install.
Connect Zep once
Add your Zep connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

Zep integration FAQ

What can you do with the Zep integration?

LiveContext exposes 18 operations on Zep, covering graph, user, thread and message. Common uses include cleaning up test conversations, auditing which conversations exist in a project and creating the memory owner when someone signs up.