Workflow recipes

tuhdoo coordinates who works what: agents claim tasks, report outcomes, and escalate questions through a small MCP surface, and the ledger of all of it lives on a git branch inside your repo. What tuhdoo deliberately does not touch is how the code itself gets written and merged — the agent protocol's step for that is "ordinary git on ordinary branches", and it stays that way no matter what your team does.

That leaves a real question unanswered: what should the git workflow around an agent fleet look like? Recipes are our answer — recommended patterns for the outer workflow: branching shape, PR conventions, how work lands, and the repo settings that keep a fleet honest. The boundary is deliberate:

• Protocol ships in the tool and in agent-protocol.md — the claim → work → escalate → finish loop, lease semantics, honest outcomes. Agents must follow it; the daemon enforces the parts that can be enforced.
• Recipes are suggestions. Adopt one wholesale, adapt it, or ignore it — tuhdoo behaves identically either way.

Each recipe ends with a copy-pasteable instructions block for your repo's agent-instructions file (CLAUDE.md, AGENTS.md, or whatever your harness reads), ready to adapt.

The recipes

• trunk-based-pr-flow.md — one task, one branch, one squash-merged PR into a protected default branch. The flow tuhdoo's own repository is built with; start here.