Zep
The Foundational Memory Layer for AI. Equip your agents with the knowledge to complete tasks, from the mundane to monumental.
Verdict
Common use cases
- Recall client details from past support chats
- Track research findings across multiple sessions
- Build knowledge graphs from meeting transcripts
- Retrieve context for long-running projects
- Clone user graphs for testing workflows
Integration
- Vendor
- Zep
- Category
- other
- Auth
- API_KEY
- Tools
- 43
- Composio slug
zep
Tools
- Add Fact Triple
Tool to add a manually specified fact triple (subject-predicate-object) to the Zep knowledge graph. Use when you need to add explicit relationships between entities. Returns a task_id to monitor processing status.
- Add Session Memory
Tool to add memory messages to a specified Zep session. Use when you need to store conversation history or context in a session.
- Add Thread Messages
Tool to add chat messages to a thread in Zep and ingest them into the user knowledge graph. Use when you need to add conversation history to a thread - for best results, add messages on every chat turn in the order they were created.
- Clone Graph
Tool to clone a user or group graph with new identifiers in Zep. Use when you need to create test copies of user data, migrate user graphs to new identifiers, or set up template graphs for new users. This is an asynchronous operation that r
- Create Graph
Tool to create a new graph by adding data to Zep. Use when you need to add text, message, or JSON data to a user's graph or a specific graph. The data is processed and an episode node is created in the graph.
- Create Group
Tool to create a new group in Zep for multi-user graph management. Use when you need to create a namespace for shared context across multiple users.
- Create Session
Tool to create a new session in Zep for storing conversation memory. Use when you need to establish a new conversation context linked to an existing user. The user must be created first before creating a session.
- Create Thread
Tool to create a new thread in Zep for a specific user. Use when you need to start a new conversation thread. The user must be created first before creating a thread. Zep automatically warms the cache for that user's graph data in the backg
- Create User
Tool to create a new user in Zep with properties like user_id, email, and metadata. Use when you need to add a new user to the system. It is recommended to provide at least first_name and ideally last_name for better user association.
- Delete Graphdestructive
Tool to delete a graph from Zep. Use when you need to permanently remove a graph and all associated data.
- Delete Groupdestructive
Tool to delete a group from Zep. Use when you need to permanently remove a group and its associated data.
- Delete Session Memorydestructive
Tool to delete a session and its memory from Zep. Use when you need to permanently remove all memory data associated with a specific session.
- Delete Threaddestructive
Tool to delete a thread and its messages from Zep. Note that deleting a thread removes the thread and its messages from the thread history but does not delete associated data in the user's knowledge graph.
- Delete Userdestructive
Tool to delete a user and all associated threads and artifacts from Zep. Use when you need to permanently remove a user and handle Right To Be Forgotten requests. Deleting a user will delete all threads and thread artifacts associated with
- Get Edge by UUID
Tool to retrieve a specific edge by its UUID from the Zep knowledge graph. Use when you need to fetch detailed information about a relationship between nodes, including the semantic fact, connected nodes, and temporal metadata.
- Get Graph by ID
Tool to retrieve a graph by its unique identifier from Zep. Use when you need to fetch details about a specific graph including its name, description, and timestamps.
- Get Group by ID
Tool to retrieve a group by ID from Zep. Use when you need to fetch detailed information about a specific group including its configuration and metadata.
- Get Node Entity Edges
Tool to retrieve all entity edges for a specific node in the Zep knowledge graph. Use when you need to fetch relationship information, facts, and connections for a given node UUID.
- Get Project Info
Tool to retrieve project information based on the provided API key. Use when you need to fetch project details including UUID, name, description, and creation timestamp.
- Get Session by ID
Tool to retrieve a session by its unique identifier from Zep. Use when you need to fetch details about a specific session including user association, timestamps, classifications, and metadata.
- Get Session Memory
Tool to retrieve memory for a given session including relevant facts and entities. Use when you need contextual information and historical data from a session.
- Get Session Message by UUID
Tool to retrieve a specific message by UUID from a Zep session. Use when you need to fetch a single message's details by its unique identifier from a particular session.
- Get Session Messages
Tool to retrieve messages for a given session from Zep. Use when you need to fetch the message history for a specific session with optional pagination support.
- Get Task Status
Tool to check the status of asynchronous operations in Zep. Use when monitoring batch adds, clone operations, or fact triple additions. Returns comprehensive task information including status, progress, timestamps, and error details if appl
- Get Thread Messages
Tool to retrieve conversation history for a specific thread from Zep. Use when you need to fetch the chat message history with optional pagination support via limit, cursor, or lastn parameters.
- Get Thread User Context
Tool to retrieve the most relevant user context from the user graph based on thread messages. Use when you need to get context including memory from past threads that is most relevant to the current thread.
- Get User by ID
Tool to retrieve a user by their user ID from Zep. Use when you need to fetch detailed information about a specific user including their profile, metadata, and configuration settings.
- Get User Node
Tool to retrieve a user's graph node and summary from Zep. Use when you need to access the user summary generated from instructions, build custom context blocks, or retrieve facts and information associated with a specific user.
- Get User Nodes
Tool to retrieve all nodes for a specific user from their graph in Zep. Use when you need to fetch entity information, preferences, and knowledge graph data for a user. Supports pagination via limit and uuid_cursor parameters.
- Get User Sessions
Tool to retrieve all sessions for a user from Zep. Use when you need to fetch session history for a specific user ID. Returns an array of session objects with metadata, classifications, and timestamps.
- Get User Threads
Tool to retrieve all threads for a specific user from Zep. Use when you need to fetch thread history for a specific user ID. Returns an array of thread objects with identifiers and timestamps.
- Graph Search
Tool to perform hybrid graph search combining semantic similarity and BM25 full-text search across the Zep knowledge graph. Use when you need to search for entities, relationships, or episodes within a user, group, or specific graph. Suppor
- List All Threads
Tool to list all threads with pagination and ordering support. Use when you need to retrieve threads with optional pagination (page_number, page_size) and ordering (order_by, asc) parameters.
- List Graphs
Tool to retrieve all graphs from Zep with pagination support. Use when you need to fetch a list of graphs with optional pagination via page_number and page_size parameters.
- List Groups Ordered
Tool to retrieve all groups from Zep with pagination support. Use when you need to fetch a list of groups with optional pagination via pageNumber and pageSize parameters.
- List Sessions Ordered
Tool to retrieve all sessions from Zep with pagination and ordering support. Use when you need to fetch a list of sessions with optional pagination via page_number and page_size parameters.
- List Threads
Tool to retrieve all threads from Zep with pagination support. Use when you need to fetch a list of threads with optional pagination and sorting via page_number, page_size, order_by, and asc parameters.
- List Users Ordered
Tool to retrieve all users from Zep with pagination support. Use when you need to fetch a list of users with optional pagination via pageNumber and pageSize parameters.
- Update Graph
Tool to update graph information in Zep including name and description. Use when you need to modify graph properties after creation.
- Update Group
Tool to update group information in Zep including name, description, and fact rating instructions. Use when you need to modify an existing group's properties.
- Update Message
Tool to update a message in a Zep thread. Use when you need to modify message content, metadata, role, or other properties of an existing message. Particularly useful for adding or modifying metadata after a message has been created.
- Update Session Metadata
Tool to update session metadata in Zep. Use when you need to modify or add metadata to an existing session. Metadata is merged, so existing keys are preserved unless explicitly overwritten.
- Update User
Tool to update an existing user's information in Zep including email, metadata, and ontology settings. Use when you need to modify user properties after creation.
Setup
Setup guide
- 11. Go to Settings > Integrations in Switchy and click Add Integration. 2. Select Zep from the list and choose API Key as the auth method. 3. Log into your Zep dashboard at app.getzep.com (or your self-hosted URL) and navigate to API Keys under your project settings. 4. Generate a new API key and copy it. 5. Paste the key into Switchy's integration form and click Connect. 6. Open any Space and type '@Zep create a new session for user test-user-123' to verify the connection works. 7. If the session is created successfully, Zep is ready to use. 8. To invoke memory retrieval in a Space, @mention Zep and ask it to search a session or add facts to a graph. 9. Each user or project in your team should map to a unique user ID in Zep so memories don't collide.
What teammates see: by default, memories from Zep are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Spaces have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Create User Session
@Zep create a new session for user client-acme-corp and confirm it's ready to store messages.Open in a Space →
Add Conversation Memory
@Zep add the last 10 messages from this Space to session client-acme-corp so we can reference them next week.Open in a Space →
Search Session History
@Zep search session client-acme-corp for any mentions of pricing or contract terms from our last three calls.Open in a Space →
Add Fact to Graph
@Zep add a fact triple to the graph: subject 'Acme Corp', predicate 'uses', object 'Salesforce CRM'.Open in a Space →
Clone User Graph
@Zep clone the graph for user client-acme-corp to a new identifier client-acme-corp-test for sandbox testing.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@zep search my knowledge graph for all facts about the Q4 product roadmap we discussed last month
Based on your knowledge graph, here are the key facts about Q4 product roadmap from your conversations: You prioritized mobile app redesign (discussed Dec 3), planned API v3 launch for October (mentioned in planning session Dec 1), and identified performance optimization as a blocker (flagged Dec 5). The graph also shows connections between 'mobile redesign' and 'user retention goals' with a confidence score of 0.87. Three related episodes contain additional context about timeline dependencies.
This example showcases Zep's core strength: semantic search across conversational memory stored in a knowledge graph. The MCP retrieves facts with temporal context and relationship confidence scores. Requires an existing user graph with ingested conversation history—empty graphs return no results. Search quality improves as more conversations are added to the graph over time.
@zep add this conversation to my thread and create a fact triple linking 'Sarah Chen' as 'reports to' 'Engineering VP'
I've added the current conversation messages to your thread (thread_id: th_8x2k9m) and ingested them into your knowledge graph. Additionally, I created an explicit fact triple: Subject='Sarah Chen', Predicate='reports to', Object='Engineering VP'. This relationship is now permanently stored in your graph with high confidence. You can query organizational structure or Sarah's context in future conversations. Task ID fact_task_7n3p monitors the ingestion—typically completes in 2-3 seconds.
Demonstrates Zep's dual capability: automatic conversation ingestion plus manual fact assertion. The fact triple creates a structured relationship that persists across sessions, useful for org charts, project dependencies, or domain knowledge. Requires valid thread and user IDs. Be aware that manual triples override conflicting auto-extracted facts, so use deliberately to encode ground truth your team relies on.
@zep summarize everything in my session memory related to the API migration project, then tell me what's still unresolved
From your session memory, the API migration project spans 8 conversations over 3 weeks. Key decisions: migrated authentication to OAuth2 (resolved Dec 10), chose PostgreSQL for new schema (resolved Dec 12), and set cutover date for Jan 15 (confirmed Dec 18). Unresolved items: database connection pooling strategy (last discussed Dec 14, no conclusion), rollback procedure for partial failures (raised Dec 16, awaiting DevOps input), and monitoring dashboard requirements (mentioned Dec 11, needs PM review). Your graph shows 'connection pooling' linked to 'performance concerns' with moderate confidence.
This synthesis example pairs Zep's session memory retrieval with AI reasoning to extract project status. Sessions store chronological conversation history, while the knowledge graph surfaces semantic relationships. Effective for project continuity across long timelines. Note that session memory has retention limits (check your Zep plan)—older conversations may age out. For critical long-term context, explicitly add facts to the graph rather than relying solely on session storage.
Use-case deep-dives
When Zep wins for multi-agent support escalation
A 6-person support team routes tickets across Tier 1 chat, Tier 2 email, and a product specialist. Each agent uses a different AI tool, but the customer's full history lives in Zep's session memory. When a ticket escalates, the next agent's AI pulls the session ID and loads the entire conversation graph—no copy-paste, no "what did the last person say" confusion. Zep's 43 tools handle session creation, message threading, and fact extraction without the team writing custom memory logic. This works if your escalation rate is high enough to justify the API key overhead and you're already using multiple AI agents. If your team uses one shared inbox and one AI assistant, session memory is overkill—just keep context in the thread. Buy Zep when handoffs between agents happen daily and losing context costs you repeat questions.
When Zep's knowledge graph beats CRM notes
A 4-person sales team closes deals over 8-12 touchpoints spanning calls, emails, and Slack DMs. Each rep's AI assistant writes to a Zep user graph tied to the prospect's account. The graph auto-extracts entities (budget mentioned, decision-maker titles, competitor names) and links them as fact triples. When the AE preps for a follow-up, their AI queries the graph and surfaces "CFO said Q3 budget unlocks" without the rep digging through CRM activity logs. Zep's Add Fact Triple and Create Graph tools build this automatically if your team logs conversations in AI tools already. The threshold: if your deal cycle is under 3 touches or your CRM search is fast enough, the graph is extra complexity. Buy Zep when your team closes complex deals and reps waste 20+ minutes per week hunting old context.
When Zep's group graphs speed up new hire ramp
A 10-person engineering team onboards 2-3 developers per quarter. Each new hire's AI assistant joins a Zep group graph seeded with the team's architecture decisions, incident post-mortems, and code review patterns. As the hire asks questions, their assistant queries the group graph and returns answers grounded in the team's actual history—not generic docs. Senior engineers add facts to the graph when they explain a gnarly design choice, and those explanations become searchable for the next hire. Zep's Create Group and Clone Graph tools let you template this setup once and replicate it per cohort. This pays off if onboarding takes over 4 weeks and your team documents decisions in chat or tickets. If your team is under 5 people or onboarding is ad-hoc, a shared wiki is simpler. Buy Zep when new hires ask the same 30 questions every quarter and senior time is your bottleneck.
Frequently asked
What does the Zep MCP do in Switchy?
Zep MCP turns your team's conversations into a queryable knowledge graph. It stores chat history, extracts facts and relationships between entities, and lets AI agents recall context across sessions. Think of it as long-term memory for your AI workspace — agents can reference what happened three weeks ago without re-reading transcripts.
Do I need a Zep account to use this MCP?
Yes. You need a Zep API key, which means you need a Zep account (free or paid). Paste the key into Switchy's MCP settings. No OAuth dance — just API key auth. Anyone on your Switchy team with the key can connect it, but you probably want one shared Zep project per team to avoid fragmenting your knowledge graph.
Can Zep MCP search my existing app data like Notion or Slack?
No. Zep doesn't connect to third-party apps. It only stores data you explicitly send it — chat messages, facts, JSON blobs. If you want Slack history in Zep, you'd pull it via Slack MCP first, then push it to Zep. Zep is a memory layer, not a data ingestion platform.
How is this different from just storing chat logs in a database?
Zep builds a semantic graph, not a flat log. It extracts entities, relationships, and facts from conversations, then lets you query by meaning — "what did we decide about the pricing model?" — instead of keyword search. The graph also powers context-aware retrieval, so agents surface relevant history without scanning every message.
Who on the team should set up the Zep MCP?
Whoever manages your AI infrastructure. You need to decide session/thread naming conventions and whether to use user graphs or group graphs. If multiple people connect Zep with different API keys, you'll fragment your knowledge graph across projects. One shared key, one graph, one source of truth.