Meko for Lyzr
Lyzr deploys a production agent for a task, but the agent’s knowledge disappears when the session is over. The next agent you spin up starts from zero. Lyzr Agent Studio supports custom MCP servers, and Meko is a remote MCP server, so the tool card mechanism Studio already documents works for Meko.
Over MCP, Meko provides Lyzr with:
Agent-scoped memory
What your agent learns persists beyond the session and is readable via every AI tool you connect to.
A shared knowledge base
Documents and promoted conclusions are searchable by all agents and teammates with access to the datapack.
A trace of every read and write
Each conversation is stored with its reasoning, so you can see what an agent did and why.
Every agent you spin up starts with what the last one learned
Lyzr Agent Studio supports custom MCP servers, and Meko is a remote MCP server, so the tool card mechanism Studio already documents works for Meko as-is.
Studio agents share one datapack, scoped per agent, with a record of what each did and why. Integration is a no-code tool card in Agent Studio; always-on behavior requires a Lyzr skill you write from the community skills repo.
How Meko Benefits Lyzr
Add Meko without code
Studio’s MCP tool type handles the connection: add the server URL and your API key, and Meko’s tools appear for your agents to use.
Grant only the actions each agent needs
Grant an agent the search, memory, and conversation tools it needs, and keep datapack administration off the grant.
Let the fleet learn together
Every agent granted the Meko tool card reads and writes the same datapack, tagged by agent_id, so what one agent learns is available when the next one starts.
Trace what each agent did
What each agent did and the reasoning behind it are readable in any MCP client on the project.
Connecting Meko in Lyzr
- In Agent Studio, go to Connections, then Tools, and add an MCP Server: name it Meko, set the Server URL to https://mcp.mekodata.ai/mcp, and provide your Meko API key (Lyzr also supports connecting with OAuth).
- In your agent’s configuration, add the Meko tool and select the actions it needs.
- In the agent’s instructions, tell it which datapack to use, just as you would with any other agent.
- For always-on record-and-recall behavior, adapt the meko-mcp-tools behavioral guide from the Meko community skills repo into a skill for your agent. Lyzr agents load skills you define in Studio, so the guide’s record and recall rules drop in as instructions.
YugabyteDB is also available out of the box as a vector store in Lyzr’s data connectors, separate from Meko. If your agent uses both a YugabyteDB-backed knowledge base and a Meko datapack, keep your uploads in one or the other to avoid maintaining two copies of the same knowledge.