Start here
Pick your path into SEIF — whether you are an AI agent, an operator running workspaces, or an engineer integrating SEIF into a system.
SEIF serves two kinds of reader at once: the human who wants durable, verifiable work, and the AI agent who has to reach the right command unaided. These pages are short maps — each one points you straight at the concepts, guides, and reference you need, in order. Pick the path that fits you.
For Hub users
You govern in the browser — invite, onboarding, Propose, Access, Explore. Honest about what still needs the CLI.
For AI agents
You landed in a SEIF workspace with no prior context. Become a just-in-time SEIF expert: ground, remember, reach the right verb, govern your work.
For operators
You run workspaces day to day. Initialize, build durable memory, open and close cycles, and keep machines in sync.
For integrators
You are wiring SEIF into another system — an editor, an agent framework, or a provenance pipeline. Connect over MCP and the CLI, then verify.
The shape of the protocol
Whichever path you take, four ideas carry the whole model: the
workspace and its .seif/ store,
modules that hold replayable knowledge,
evidence that makes work verifiable,
and the operating conduct every agent
follows. Read What is SEIF once and the rest of
the docs fall into place.