Why do some Jules PRs get rejected?

Three reasons show up again and again: vague issue text, missing tests to verify against, and repo files the agent cannot read (generated code, vendored libraries). Jules does what the repo makes easy.

How do you prepare a repo for Jules?

1. Add a JULES.md or clear README section listing build and test commands.
2. Make sure pytest or npm test runs clean on a fresh clone.
3. Keep test runtime under two minutes — slow suites stall the agent loop.
4. Exclude generated and vendor directories so the agent focuses on source.
5. Tag good starter issues with a label so you can queue them in batches.

That third point matters more than it sounds. If your suite takes ten minutes, the agent spends most of its budget waiting.

How should you frame a task?

Write tasks with an input, an expected output, and a constraint. Compare these two:

The strong version tells the agent where to put the test and what must not change. That is what keeps the diff small.

What is a realistic review workflow?

Treat a Google Jules PR exactly like an external contributor’s. Run it locally before approving. Check that the test actually fails on the old code — agents sometimes write tests that pass no matter what.

I keep a rule: no agent PR merges without a human typing the manual reproduction once. It catches the cases where the test suite is green but the bug is still there.

Jules vs other agents: when to pick it?

Pick Jules for backlog cleanup on repos you already trust. Skip it for greenfield architecture decisions or anything touching billing logic, where a wrong-but-passing change is expensive. It is a volume tool, not a design tool.

Quick Summary

• Repo prep (fast tests, build docs) drives Jules merge rates more than prompt wording.
• Write tasks with input, expected output, and a constraint.
• Keep test suites under ~2 minutes for best throughput.
• Verify that new tests actually fail on the old code.
• Use it for backlog work, not architecture or billing logic.

What is Google Jules?

Jules is Google’s answer to Devin and Cursor’s background agents. Instead of autocompleting in your editor, it runs a full clone, reads the codebase, makes edits, runs your tests, and opens a PR you review like any teammate’s.

It is powered by Gemini models and currently supports Python and JavaScript projects best. Anything with a clear test suite gets the most reliable results.

Is Google Jules free?

Jules launched with a free tier that gives a limited number of daily tasks, and Google introduced paid plans for higher task counts. The free allowance is enough to try it on real work but not to run it all day. Check the pricing page before you depend on it for a team.

How do you connect a repository?

1. Sign in to Jules with your Google account.
2. Install the Jules GitHub app on the account that owns the repo.
3. Pick the repository from the dashboard list.
4. Choose the base branch (usually main).

Jules asks for read and write access so it can push branches. If your org blocks third-party apps, you need an admin to approve it first.

How do you run your first task?

Type a task the way you would describe it to a junior developer. “Add input validation to the signup endpoint and cover it with a test” works far better than “fix the code.”

Jules returns a plan you can approve or edit before it starts. Approve it, then wait. Small tasks finish in a few minutes; bigger ones run longer, and you get a notification when the PR is ready.

What does Google Jules get right and wrong?

Good fit: maintainers drowning in small labeled issues, and teams with good test coverage. The async model means you can queue three tasks and review PRs later.

Weak fit: large monorepos with slow test suites, and anything requiring deep domain knowledge the agent cannot infer. I have seen it pass tests while missing the actual intent of an issue, so read the diff. It also cannot see private docs or your Slack context.

Quick Summary

• Google Jules is an async coding agent that opens pull requests on GitHub.
• It runs in a cloud VM and needs the Jules GitHub app installed.
• A free task allowance exists; heavy use needs a paid plan.
• Works best on Python and JavaScript repos with tests.
• Always review the diff — passing tests does not guarantee correct intent.