Nauro-AI

nauro-adopt

Seeds Nauro's project store from an existing repo. Use after `nauro adopt` has run locally. On filesystem-capable surfaces, reads docs (README, manifests, ADRs, Memory-Bank) for rationale and inspects code, config, tests, lockfiles, and recent git history for evidence, then surfaces targeted probes that turn evidence into rationale. On chat surfaces, operates on pasted content against an already-adopted project.

nauro-context

Writes durable shared context into Nauro's project store so other agents (a later session or a parallel one) can discover and pull it, finds and reads context another agent left, or captures a resumable brief so your own next session in this environment picks up cleanly. Three modes. Author writes a shared brief for any agent. Find locates and reads a brief another agent left. Resume captures a self-directed brief and hands back a short prompt to start the next session. Offer Resume mode when the user asks (in their own words) to give me a prompt for a fresh session or instance, hand off this work, or write a resume doc, and let the user accept before running it. Briefs land at <store>/context/<slug>.md (picked up by `nauro sync` with no code change); Author flags a BRIEF discovery pointer and Resume flags a RESUME pointer naming that path. Uses the agent's filesystem write and the `nauro status` shell command to resolve the store path, alongside the MCP tools get_context, get_raw_file, and flag_question; nev

nauro-loop

Originate gated Delivery and Interview candidates, or coordinate selected Program Delivery as FRAME -> CHOOSE -> START -> ADVISE -> VERIFY -> ADVANCE. Human-named work bypasses candidate selection. Agent-originated work keeps read-only ORIENT, 1-3 candidates, mandatory human selection, reject-all, and no auto-pick path. Each Program slice uses at most one fresh direct-user Delivery task. Automatic launch requires surface lifecycle support to create, identify, inspect, and message that task; otherwise the coordinator returns one exact launch prompt and stops. Coordinator artifact review is advisory, and integration is verified independently. Synchronous non-program Delivery stays outside the Program state machine. Interview stays explicit and non-authoritative. Ordinary outputs create no automatic store artifacts; scheduled ORIENT retains its existing SELECT checkpoint and pointer writes as a narrow process-state exception. Installed by `nauro adopt --with-skills`.

nauro-ship-task

Run the full planner -> executor -> reviewer -> tech-lead -> direct-user-confirm -> push chain for a non-trivial code change against Nauro's bundled @nauro-* subagents. Every subagent is draft-only for project-truth writes; the direct-user Delivery parent files exact approved decision proposals. Runs @nauro-tech-lead Mode C between reviewer-APPROVE and the push gate to catch doctrine drift the reviewer missed. A prompt that carries a detailed implementation spec or a pasted handoff is still chain input, not license to implement directly. Invoke explicitly with the surface's nauro-ship-task command. Requires `nauro adopt --with-subagents` to have run. A program Delivery returns a standardized program handback after PR creation or a terminal blocker.

nauro-interview

Ask compact, numbered prerequisite-ready questions to elicit tacit project reasoning or challenge a proposed choice against Nauro decisions and repository evidence. Continue until every material branch has a disposition, then classify the result as shared understanding without granting write authority. Use only when the user explicitly asks to be interviewed, grilled, stress-tested, or helped to transfer reasoning into Nauro. Runs in the main agent context with no external skill or subagent dependency.