Docs · hypothesis-led
Meaningful Launchfiles work maps to Signal → Decision → Work Order → Run → Gate → Approval → Receipt → Readout → Memory. A model reply or green UI status is not execution proof. High-risk actions require evidence and authority outside the agent.
Docs hubGet startedllms.txtAgent capabilities
Hypothesis-led documentation. No ranking, traffic, or outcome guarantees. Public discovery is not execution authority.
Speed without durable context creates false progress. The loop forces atomic work, independent evaluation, durable receipts, and optional compounding memory.
Side-effectful or claim-making work needs Gate and Receipt at minimum. Read-only orientation can be lighter, but anything asserted as truth still needs an inspectable receipt.
Founders usually work from /today. Runtime Map is forensics. Public marketing pages explain objects; this docs page is the contract-oriented explanation.
Next safe action
Prefer receipts over claims. Prefer fail-closed honesty over impressive incompleteness.
