docs(testing): describe how run-e2e.sh actually picks the runner

[teamleaderleo]

The cmux-testing reference said the wrapper sends a commit to the 12vcpu or 6vcpu Blacksmith pool by the parity of its SHA. That rule is gone: auto now goes through scripts/ci/e2e_runner_pool.py, the same rule pull request CI uses (an owned Mac with a free slot, then 12vcpu while it has headroom, then 6vcpu, then the shorter queue).

Agents read this file before dispatching E2E runs. It now says that naming a Blacksmith pool skips idle owned Macs, and that a pinned pool's queue can run past an hour. At 11:40 UTC today the Blacksmith macOS pools held about 237 queued jobs, with the oldest waiting 76 minutes.

Docs only.

🤖 Generated with Claude Code

---

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

---

Summary by cubic

Updates the cmux-testing docs to reflect how run-e2e.sh actually picks its runner. The old rule (runner chosen by the parity of the commit's SHA) is gone; auto now uses scripts/ci/e2e_runner_pool.py, the same logic pull request CI uses: an owned Mac with a free slot, then blacksmith-12vcpu-macos-26 while it has headroom, then blacksmith-6vcpu-macos-26, then the shorter queue. The docs also note that naming a Blacksmith pool skips idle owned Macs and can queue for an hour or more when many branches are running at once.

• Leaving --runner off also lets the wrapper attach to an identical run already in flight at that commit on any of those pools; naming a pool only reuses runs on that pool.

Docs only.

^{Written for commit c901c24. Summary will update on new commits.}

Summary by CodeRabbit

• Documentation
  • Clarified how automatic runner selection prioritizes available Macs and Blacksmith pools, and how specifying a pool affects wait times and reuse of in-flight runs.
  • Confirmed that the macOS 15 runner remains available for macOS 15-specific checks.

[github-actions]

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

[coderabbitai]

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

Note

Currently processing new changes in this PR. This may take a few minutes, please wait...

⚙️ Run configuration

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

Review profile: ASSERTIVE

Plan: Advanced

Run ID: 31386779-e9ef-4d58-9ca8-8b914042c9af

📥 Commits

Reviewing files that changed from the base of the PR and between 0dba677 and c901c24.

📒 Files selected for processing (1)

• skills/cmux-testing/references/local-vs-ci-validation.md

 ________________________________________________________________________
< I'm CodeRabbit. I review the code you never knew you needed reviewing. >
 ------------------------------------------------------------------------
  \
   \   (\__/)
       (•ㅅ•)
       / 　 づ

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Commit to this branch
• Create a new PR

---

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.}

[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.

[github-actions]

Merge receipt for c901c2452f: every check was green at merge (8 verified; 11 skipped by policy). Full suite runs on main after merge.