First task walkthrough

Quick Start gets the tool installed. Local initialization with 0dai init --local needs no account. For the next step, 0dai run "…" --now first contacts the 0dai planning service, then uses your installed agent CLI and configured provider to execute the task. Check service access and provider setup before running it; this is not an offline run. Scaling that to a branch and pull request is the team/Pro flow, covered lower down.

After local initialization, choose one small task on your own repo. When service access and your agent are ready, try 0dai run "add a test" --now. Read the result and receipt, and inspect the diff before accepting the change.

Target outcome (team/Pro: branch + PR)

Once run --now feels useful, the team/Pro swarm flow takes a tiny issue all the way to a reviewable pull request. Use a small task: add a health check endpoint, fix one failing test, or update a doc link — small enough for a junior developer to review without trusting the agent blindly.

A real repository, not a mocked transcript.
A branch with one narrow change.
Tests, lint, or an explicit note explaining why none apply.
A pull request with the command run and the verification result.

Commands

curl -fsSL https://0dai.dev/install.sh | sh
cd your-project
0dai init --local
0dai run "add a test for the auth module" --now

0dai run --now uses the planning service before local execution with your installed agent CLI (claude, codex, …) and its configured provider. Model-provider usage and your host are separate from your 0dai plan. Run 0dai doctor to check which agent CLIs are available. The branch-and-pull-request flow below is the team/Pro swarm path and also needs the gh CLI.

What to watch for

The useful part of the walkthrough is not that an agent typed code. It is the control loop around the code: the agent reads the project layer, scopes the task, writes a small patch, verifies it, and leaves a receipt.

Moment	Good sign	Stop if
`0dai init --local`	`ai/` and native agent configs are created.	The command wants secrets, production credentials, or broad repo access.
`0dai doctor`	It reports the project layer and agent config health.	It cannot find the repo root or generated config files.
First task	The branch touches only files needed for the issue.	The agent rewrites unrelated files or expands the goal.
Pull request	The PR body names the issue, files touched, and checks run.	The PR claims a test passed without a command or log.

PR review receipt (team/Pro)

Distinct from the run --now scored receipt above: when you scale to the team/Pro swarm flow, the pull request itself is the receipt. A good first PR body is short and specific — enough context for a reviewer to decide whether the change is narrow.

Fixes #123

Changed:
- added GET /health
- added route test

Verified:
- npm test -- health
- npm run lint

Video publish gate

The landing-page video should publish only after these are true:

It shows a real repository and a real pull request.
It includes faces and voices or a spoken walkthrough, not only a silent screencast.
Captions are present and accurate.
No metric, time, or customer claim is invented for the recording.
The final cut stays under 10 minutes and links back to this page.

Next up

Setup guide - install and account activation.
Multi-agent workflow - how delegation works after the first task.
Session roaming - hand off context between agents.