Solutions
Seven primitives. Every kind of agent.
Coding agents, support bots and research swarms all need the same things: somewhere isolated to work, state that survives, and credentials they can use but never read. Here is how teams put the primitives together.
Use cases
Coding agents
A filesystem, database branch and scoped credentials for every task.
Multi-agent systems
Queues, shared memory and forked workspaces for agents that work together.
Long-term memory
Typed memory that persists across sessions and keeps its history.
Agent platforms
An isolated group per customer, scaled to zero when idle.
Evals and experiments
Fork the same starting state for every run and compare the results.
Self-hosted and local
Run the open-source engine on your laptop, in CI or in your own cloud.
Industries
Developer platforms
Backends for the apps and agents your users build.
Customer support
Support agents that remember each customer and follow up on time.
Financial services
Isolated client data, credentials agents can't read and a full audit trail.
Security operations
Isolated workspaces for investigations, with tools locked down by policy.
Retail and e-commerce
Shopping agents that remember preferences and act on store APIs safely.
Education
Tutors that remember each learner, with a workspace per student.
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