Use case · Agent builders
Memory and tools for the agents you build
Point your own agent at Memtro over the OpenAI-compatible API or MCP and it gets long-term memory, read-only connectors and model routing without you building any of it.
The problem
- Every agent reinvents memory, and none of them share it
- Giving an agent safe access to company systems is a project in itself
- Tool sprawl: each framework wants its own connectors
How Memtro helps
- One API key per user carries their memory, their connections and their permissions, whatever client is asking
- memtro_search, memtro_remember and memtro_explore give any MCP client the same memory the dashboard uses
- Model routing picks a rung per request, so the agent is cheap for easy turns and strong for hard ones
Example queries
- › Search memtro for what we know about Acme before drafting the proposal
- › Remember that the client prefers invoices on the 1st of the month
- › Use memtro to pull the open tickets for this account and summarise them
Jobs to schedule
- schedule: Every night, re-run the research agent for accounts with a renewal in 30 days and save the brief to memory
- schedule: Every hour, let the triage agent check new tickets and remember anything that needs engineering
Say any of these in Chat and Memtro creates the job. See Jobs.