cd /news/ai-agents/the-best-onboarding-flow-is-a-pull-r… · home topics ai-agents article
[ARTICLE · art-19467] src=charlielabs.ai pub= topic=ai-agents verified=true sentiment=↑ positive

The Best Onboarding Flow Is a Pull Request

Charlie Labs has introduced a new onboarding method for its Daemon software that uses a pull request instead of a traditional setup wizard. The process begins with a single, high-value task from a catalog of common engineering jobs, such as managing pull request metadata or dependency updates, and expands only after the team reviews and approves the results. This approach ensures engineers understand the software's role and boundaries before it operates autonomously within their repositories.

read2 min publishedJun 2, 2026

Claude just discovered workflows. Charlie started there.

Claude just discovered workflows. Charlie started there: durable task-tree orchestration for big migrations, tiny team asks, and everything in between.

Start with one useful job, review the setup PR, and expand only after the results are clear.

Most software onboarding confuses completion with value. Create an account. Connect every integration. Click “next” six times. Land in an empty dashboard. The product tour is over; the customer is still on their own.

That model works poorly for software that will operate inside an engineering team. Access is necessary, but access alone is not a role. If software will take initiative around repositories, issues, or team workflows, engineers need to understand what job it owns, when it should act, where its work will appear, and what it must not do.

That is why the best onboarding flow for a Daemon looks less like a setup wizard and more like a pull request.

The idea applies beyond Charlie. Agentic software should start with one useful job, make the operating policy easy to inspect, and expand only after the team sees useful results.

For Charlie, the quickest path is to start with a simple, high-value Daemon from the [catalog](https://charlielabs.ai/use-cases/). The catalog covers common under-owned work such as pull request metadata, failing checks, stale docs, dependency updates, and Linear issue hygiene. Pick the narrow role that removes a real maintenance burden for the team.

For a Daemon, the onboarding flow is short:

`.agents/daemons/<daemon-id>/DAEMON.md`

for the repo, and opens a setup pull request.The setup PR gives the team something concrete to inspect, merge, observe, and revise. Trust grows from a role that wakes for the right reasons and leaves reviewable work behind.

Agents create work. Daemons do the rest.

Ask Charlie to recommend one simple catalog Daemon for your project and open the setup PR. If Charlie is not connected to the repository yet, install the GitHub App: https://docs.charlielabs.ai/installation

── more in #ai-agents 4 stories · sorted by recency
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/the-best-onboarding-…] indexed:0 read:2min 2026-06-02 ·