Writes thread messages into a user graph, retrieves the context relevant to the conversation, searches it semantically, and deletes data on request.
43actions available
Every action runs live through MCP. Nothing to build, nothing to maintain.
Ayro stores every customer message in Zep's conversation graph to build a complete memory. Your team gives Ayro full context of the entire customer relationship.
Ayro retrieves the most relevant facts about a customer from their knowledge graph. Your team personalizes responses by pulling in what Ayro already knows about them.
Ayro starts a new conversation thread for each customer to organize their history. Your team keeps conversations tidy and retrieves past interactions instantly.
Each card is one request a support team gets, and the Zep actions Ayro runs to close it.
Ayro pulls the most relevant context from a user's graph based on the thread in progress, together with the facts and entities held on the session and the summary stored on that user's node.
Ayro adds chat messages to a thread and ingests them into the user's knowledge graph, and records an explicit subject-predicate-object fact when something should be stated outright rather than inferred.
Ayro creates the user record a thread has to hang off, opens the thread itself, and sets up a group namespace when several users need to draw on one shared body of context.
Ayro runs a hybrid search combining semantic similarity with BM25 full-text matching across the knowledge graph, then walks the entity edges around a matched node to see how those facts relate.
How it works
Connect once, then just ask. There is no workflow builder to learn and nothing to maintain — Ayro reads the Zep actions it has and picks the ones a request needs.
Authorize the Zep account your team already uses — one consent screen, no API keys, no mapping tables. Ayro can only do what you granted that account, and you can disconnect it at any time.
Describe the job the way you would hand it to a teammate. Ayro maps it to the Zep actions that close it and chains as many as the request needs.
The agent runs the actions inside the conversation the customer is already in, so nobody copies data between tabs and your team can take over at any point.
Get started freeZep · Ayro
What can you do in Zep without me setting anything up?
These, right away:
Great, let's start with the first one.
FAQ
Setup, permissions, and the limits of what Ayro can do inside Zep.
Get Task Status checks the state of Zep's asynchronous operations - batch adds, graph clones, and fact triple additions among them - so Ayro can confirm the write landed before relying on it, rather than reading a graph that is still ingesting.
All 43 actions your agent can call on Zep, straight from the live MCP connection.
Add a manually specified fact triple (subject-predicate-object) to the Zep knowledge graph.
Add memory messages to a specified Zep session.
Add chat messages to a thread in Zep and ingest them into the user knowledge graph.
Clone a user or group graph with new identifiers in Zep.
Create a new graph by adding data to Zep.
Create a new group in Zep for multi-user graph management.
Create a new session in Zep for storing conversation memory.
Create a new thread in Zep for a specific user.
Create a new user in Zep with properties like user_id, email, and metadata.
Delete a graph from Zep.
Delete a group from Zep.
Delete a session and its memory from Zep.
Delete a thread and its messages from Zep.
Delete a user and all associated threads and artifacts from Zep.
Retrieve a specific edge by its UUID from the Zep knowledge graph.
Retrieve a graph by its unique identifier from Zep.
Retrieve a group by ID from Zep.
Retrieve all entity edges for a specific node in the Zep knowledge graph.
Retrieve project information based on the provided API key.
Retrieve a session by its unique identifier from Zep.
Retrieve memory for a given session including relevant facts and entities.
Retrieve a specific message by UUID from a Zep session.
Retrieve messages for a given session from Zep.
Check the status of asynchronous operations in Zep.
Retrieve conversation history for a specific thread from Zep.
Retrieve the most relevant user context from the user graph based on thread messages.
Retrieve a user by their user ID from Zep.
Retrieve a user's graph node and summary from Zep.
Retrieve all nodes for a specific user from their graph in Zep.
Retrieve all sessions for a user from Zep.
Same connection, same setup. Pick the next one your team already uses.
Attributes workspace usage to individual bots, provisions workspaces and integrations, updates workflows, and drives runtime conversations.
Registers a GitHub repository for analysis and answers plain-language questions about its architecture, functionality, and dependencies.
Searches datasets, models, and competitions, publishes and versions datasets, pulls notebook code and output, and files competition entries.
Route a request to a self-hosted model, reuse OpenAI-format endpoints against it, and report which models the host currently has installed.
Creates image, video, LiDAR, and text annotation tasks, batches and finalizes them, reports completion counts, and reassigns Studio projects.
Scores a submitted passage for AI authorship, checks it for plagiarism against online sources, and measures how closely two texts match.

Authorize the account and your agent has all 43 actions from the first conversation.
