Use case · Multi-agent systems
Hand off work, not context.
Agents coordinate through durable queues and shared memory instead of passing transcripts around. Each agent gets only the tools its role needs.
How it fits
Built from the primitives you already get.
Durable handoffs
One agent sends a job, another leases it. Heartbeats, retries, delays and dead letters keep long-running work moving when an agent crashes.
Shared memory
Put a memory in the group and every agent can recall what the others learned, filtered by kind and tags.
Isolated sub-agents
Fork a directory for each sub-agent so parallel work never collides, then keep the result you want.
Least privilege per role
Limit each agent's key to the tools its role needs, such as read-only db_query for a reviewer. Enforced at the primitive, not in the prompt.
Build
Local-first development.
Build and test on your machine with the open-source engine. No cloud account needed: every primitive runs locally, with the same API as the cloud.
Install the CLI, then run primitives dev. Quickstart
Open source
Built in the open. Contribute freely.
The SDKs, the MCP server and the single-node engine are open source and actively maintained. Submit PRs, report issues, improve docs, or just star the repo to show support.
useprimitives/primitivesFork, star, or open a pull request