docs: tell contributors to start after claiming and not wait on app-host tests

[teamleaderleo]

Summary

Contributors have been claiming help wanted issues and then waiting for a go-ahead that never comes (#15710, #15743, #15755 got no reply), and one stepped back from #15752 because they couldn't run the app-host tests locally. Nothing in the entry docs said either was fine.

• docs/start-here.md: claim with one comment, then start; no need to wait for a reply. Skip issues labeled needs a call. Say so on the issue if you stop. Don't let app-host or UI tests block the PR, since CI runs the app-host suites your diff touches. Notes that the CLA check matches commit emails (fix(cli): reject trailing auth arguments #15805 is stuck on that). Picks up the squash-merge and fix-forward notes from CONTRIBUTING.
• CONTRIBUTING.md: cut to the mechanics (274 lines changed, about 90 shorter). Setup is one code block plus notes; "Running Tests" and "What CI runs for you" are one section with the app-host note up front; the PR section is one checklist. The "what a good PR looks like" and "how review works" prose it duplicated from start-here is gone, with a link. Every anchor other docs link to (#getting-started, #fast-checks-before-building-or-pushing) is kept.
• docs/contributor-verification.md: a missing local app-host run isn't a reason to hold back the PR.
• .github/contributor/welcome.md: the same note in the first-PR welcome.

Testing

• python3 scripts/verify-local.py --affected upstream/main: 15/15 selected checks passed.
• Checked every in-repo link into CONTRIBUTING.md and start-here.md still resolves to a heading or anchor that exists.
• Docs only; no build.

Changelog

none

🤖 Generated with Claude Code

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

---

Summary by cubic

Clarifies the contributor docs so new contributors start working as soon as they claim a help wanted issue instead of waiting for a reply that never comes, and so missing local app-host tests no longer block a pull request.

• Tells contributors to claim an issue in one comment, then start, and to skip issues labeled needs a call.
• Says a missing app-host or UI test run isn't a reason to hold back the pull request, since CI runs the suites the diff touches.
• Notes that the CLA check matches commit author emails to GitHub accounts.
• Trims CONTRIBUTING.md to the mechanics and removes the review and PR-quality prose it duplicated from start-here.md, keeping the biome check . scoping note and anchors other docs link to.

^{Written for commit 04483d9. Summary will update on new commits.}

[coderabbitai]

Warning

Review limit reached

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Next included review available in 1 minute.

Check out review usage here.

View limit details

Limit details: You’ve used all 10 included reviews currently available.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: Repository: manaflow-ai/cmux/.coderabbit.yaml

Review profile: ASSERTIVE

Plan: Advanced

Run ID: de724eea-5968-441d-b0df-b817553168ae

📥 Commits

Reviewing files that changed from the base of the PR and between b3a1ca1 and 04483d9.

📒 Files selected for processing (4)

• .github/contributor/welcome.md
• CONTRIBUTING.md
• docs/contributor-verification.md
• docs/start-here.md

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Autopilot is currently an internal CodeRabbit preview.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[github-actions]

CI failure attribution

CI passes on 4468056840 (run 36739173824 attempt 1).

Written by scripts/ci/classify_failures.py (ci-failure-attribution.yml); signatures are its SIGNATURES table. A machine verdict is the runner's fault, not this PR's.

[github-actions]

Dogfood tours of 3474a408

sidebar-and-chrome-tour at 3474a408: not run

skipped: CI left no app build for this head (its compile failed or was cancelled)

_{Tours are picked by the paths globs in dogfood/scenarios/*.json; a Dogfood-tours: a, b line in the description picks them instead (none turns this off). Look at every frame before merging: a green tour only means no step failed.}

[teamleaderleo]

/catch-up

[github-actions]

The merge with main (6d7ad149121a) is clean, but I was not allowed to push to docs/contributor-entry-refresh. Nothing changed on the branch; merge main yourself this time.

_{Catch-up run · RFC #14631}

[teamleaderleo]

/catch-up

Red here is not this PR's fault. This branch pins vendor/bonsplit at a commit that predates the sizing code on main, so macOS compile admission fails on symbols the PR never touches:

Sources/TerminalSizeBoundsOverlayView.swift:19:34: error: cannot find type 'BonsplitContrastPalette' in scope
Sources/TerminalSharingDisplay.swift:102:27: error: cannot find type 'TabPresence' in scope
** TEST BUILD FAILED **

macos / macOS status and tests then fail on their routing-check steps, which is them reporting that failure rather than finding their own. Main already restored the pin in "fix: restore the bonsplit pin main's sizing code needs" (#15930), so catching up on main is the whole fix.

— Raindrop g2 🫧 / Run: run_worker_20260930_3fc64ba6

[teamleaderleo]

/catch-up

My earlier catch-up comment on this PR did nothing: I wrapped the command in backticks, and the gate is startsWith(github.event.comment.body, '/catch-up'), so every run it started was skipped. Same mistake on 21 other PRs. This one is the real request; the reasoning is in the comment above.

— Raindrop g2 🫧 / Run: run_worker_20260930_3fc64ba6

[cursor]

Bugbot is paused — on-demand spend limit reached

Bugbot uses usage-based billing for this team and has hit its on-demand spend limit.

A team admin can raise the spend limit in the Cursor dashboard, or wait for the next billing cycle to continue.

[teamleaderleo]

Review: land. I checked what the rewrite dropped rather than what it reads like, since a 174-line deletion in a contributor entry point is mostly a question of where things went.

Re-homed correctly, verified in the new tree: the review-and-merge sequence, squash merges and fix-forward are in docs/start-here.md#5-what-happens-next and the anchor exists; "do not @-mention the review bots" survives there in new wording; the label meanings and "say on the issue that you are picking it up" are in start-here section 1, which still links docs/triage.md; the ladder link and "a parse or a build is not a test run" both survive. Additions that are worth having: app-host and UI tests explicitly not being a contributor's problem, in all three of CONTRIBUTING, start-here and .github/contributor/welcome.md; the needs a call label named where a patch can actually wait on it (the label exists and its description points at #13742); and the CLA note that the check matches commit author emails to the GitHub account, which is the thing that actually trips people.

Fixed: one item had no new home. The rewrite dropped why biome check . is scoped, and the standing constraint that it must not become a required CI check while formatting and import sorting are off. Nothing else in the tree carried it, so it is back beside the command in 04483d9985b.

Left, worth knowing but not worth holding docs on: docs/start-here.md#2-things-you-can-fix-without-a-mac-build is now load-bearing from CONTRIBUTING's prerequisites, so that heading cannot be renamed without breaking the link, and the same is true of #5-what-happens-next.

Merging on green.

— Raindrop g2 🫧

[github-actions]

Merge receipt for 04483d9985, merged 2026-09-30 21:26:10 UTC

• Not verified at merge: ci-status (not reported), macOS compile admission (in progress)
• Verified: CI fast guards, Fast static checks, GhosttyKit release check, guards (19), linux-preflight, macOS admission gate, Testbox broker trust boundary, Web complexity, web-validation
• Skipped by policy: admission-placement, browser, Claude wrapper regressions, Dogfood build #​${{ github.event.pull_request.number }}, full-suite-coverage, remote-daemon, suite-coverage, swift-package-tests, ui-tests, web, web-build, web-database-tests, and 1 more
• Full suite: runs on main after merge.

Labeled merged-unverified: if main breaks near this merge, look here first.