Use cases
Use Orcho where agent work has to become delivery.
The highest-leverage use cases are not autocomplete. They are the moments where multiple systems, tools, workers or reviewers have to converge into one accountable delivery decision.
One feature rarely belongs to one repo.
Modern product work crosses backend, web, analytics, infrastructure and review. Orcho treats the feature intent as the shared unit for cross-repo AI agent delivery, then projects it into scoped project runs that can be reviewed and accepted together.
/use-cases/cross-system-delivery/The worker can finish before the work is ready.
Agent output needs a gate that can say no. Orcho puts review inside the run: findings are typed, rejection changes control flow, repair resumes the same record, and final acceptance reads the evidence instead of trusting the worker's last message.
/use-cases/review-gates/A paused agent run should not become a restart.
Long-running agent work needs recovery as a first-class path. Orcho treats pause, halt, resume and handoff as run states, so the next worker or human can continue from the plan, event history and evidence already on record.
/use-cases/recoverable-runs/Start with one task
See the protocol work before reading every layer.
Install the public alpha for a self-serve run, or open the proof hub and inspect why a delivery was approved or blocked.