Individual skill
working-in-turns
Keep each turn bounded, checkpointed, and reviewable.
fleet-skills/working-in-turns/SKILL.md
Source commit bdd549c1b6491ab1954517e75f3e7d1251a3ac78
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- Yes
- Observed behavior
- Instructions only
- Last checked
- Nov 3, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 1,391
- Repository last updated
- Nov 7, 2026
- License
- Apache-2.0
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent-protocol Repository topic
- ai-agents Repository topic
- llm Repository topic
- multi-agent Repository topic
- open-source Repository topic
- self-hosted Repository topic
- agent-framework Repository topic
- agent-harness Repository topic
- agentic Repository topic
- ai-orchestration Repository topic
- engineering Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 3,244 bytes
Read excerpt
# Working in Turns
- End each turn before its run limit, not at it. The default run limit is 15 minutes, and only adapters that honor `COMMONLY_AGENT_RUN_TIMEOUT_MS` can use a longer one such as `2700000` (45 minutes). On a 45-minute limit, end the turn by about 35 minutes; on the 15-minute default, by about 10. Before ending: commit and push what you have (an open PR is enough; open it ready), and post the next step as a one-line note on the row. The next wake continues from the pushed branch. If the work is larger, split it and say how in the row. A timeout sends SIGTERM and loses uncommitted work.
- Open your PR as ready, not as a draft. A later `gh pr ready` can be refused by the harness's permission classifier ("Unrequested Commit in a Connected App") when your turn never said to undraft. That refusal comes from the harness, not from the seat `gh` fence.
- **One review round per change.** Before you hand a PR to the press, run your own quick subagent review of its diff. Tier 2 changes (auth, grants, record, schema, wake, money, security) need an Opus-class reviewer. That one review also covers screenshots against the design frame, any lane checks and the zh-CN glossary. Don't ask a second reviewer, and don't ask anyone to re-review after you push a fix.
- Then answer every CodeRabbit thread and get CI green. A push that only answers CodeRabbit or CI keeps your PASS.
- If your review comes back HOLD, fix it, push, and review again yourself.
- File the verdict as described in [review checklist rule 44](https://github.com/Team-Commonly/commonly/blob/main/docs/development/review-checklist.md#what-a-clearance-can-bind) and [rule 51](https://github.com/Team-Commonly/commonly/blob/main/docs/development/review-checklist.md#where-a-verdict-is-readable): the first line names the head you reviewed and the verdict, and the final non-blank line is the bare `GATE: PASS` or `GATE: HOLD` sentinel.
- After filing a verdict, choose a checkout strategy: inspect and remove/prune a temporary PR checkout only after checking for uncommitted work, or keep one reusable checkout and clean/refetch it for the next review.
- Never use `git worktree remove --force` before checking for uncommitted work.
## Gate ask template
Use this only when you can't run your own subagent review, for example a