session-flow: spawned workers stall waiting on their own background work
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 4/5
- 预计耗时
- 3-5 天
- 新手友好度
- 38/100
- Issue 类型
- 缺陷
- 描述清晰度
- 基本清楚
- 活跃度
- 活跃
- 技术栈
- shell
调研方向
Start from /session-flow:orchestrate imperative 4 and how orchestrate worker builds briefs, then find where dispatching skills inject waiting/background-job rules. Trace the harness notice "waiting on its own background work" and any coordinator SendMessage/poll path. Done when workers always get a bounded foreground-poll (or auto-nudge) so they do not end a turn waiting on their own background job, without relying on the coordinator to paste the rule.
由索引模型根据 Issue 内容生成。
描述
Problem
In the 2026-10-04 CI-performance session, spawned general-purpose workers ended their turn to "wait on background work" five separate times: a flaky-test fixer, the standards merge driver, the ci-workflows/medley driver, and the dotfiles/provisioning worker (twice). The harness then reported "waiting on its own background work". Nothing notifies the worker when that work finishes, so each one sat idle until the coordinator sent a SendMessage telling it to poll in the foreground.
/session-flow:orchestrate imperative 4 already says a background command is not a wait. But that text reaches a worker only if the coordinator pastes it into the brief, and these briefs did not carry it.
Ask
Make the rule reach workers without depending on the coordinator's memory. For example:
- a short "waiting" clause that
orchestrate workerexport always includes and that dispatching skills inject into every brief: poll in the foreground with a bounded loop, or return what you have; never end the turn to wait on your own background job; - and/or a coordinator-side check that turns the harness's "waiting on its own background work" notice into an automatic poll-in-foreground nudge.
Evidence
Session dba79e68-af19-4ecc-a0ce-69d8b1dbefbb: five task notifications reading "This agent has not reported yet: it is waiting on its own background work", each cleared by a manual SendMessage.
- 主要语言
- Shell
- 星标
- 22
- 派生
- 2
- 平均合并
- 5 小时 11 分钟
- 30 天内合并 PR
- 838
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 没有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
melodic-software/claude-code-plugins 的其他 Issue
-
good first issue needs-triage priority: medium
难度 2/5 1-3 小时 新手友好度 72/100
melodic-software/claude-code-plugins#6631 · 1 条评论 ·
维护者通常 1 天内回复
-
needs-triage
难度 2/5 1-3 小时 新手友好度 78/100
melodic-software/claude-code-plugins#6547 ·
维护者通常 1 天内回复
-
needs-triage
难度 2/5 1-3 小时 新手友好度 76/100
melodic-software/claude-code-plugins#6535 ·
维护者通常 1 天内回复
-
test_comment_census.py: SccArgv flag-shaped-filename test errors on Windows (#!/bin/sh scc shim)未关闭good first issue needs-triage priority: low
难度 2/5 1-3 小时 新手友好度 72/100
melodic-software/claude-code-plugins#6532 · 1 条评论 ·
维护者通常 1 天内回复
-
good first issue needs-triage priority: low
难度 2/5 1-3 小时 新手友好度 72/100
melodic-software/claude-code-plugins#6390 · 1 条评论 ·
维护者通常 1 天内回复
查看 melodic-software/claude-code-plugins 的全部 Issue
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 77/100
FluidNumerics/fluid-walk-blocker#201 ·
维护者通常 1 天内回复
-
cao-evolution cao-evolution:compiler-security
难度 1/5 1-3 小时 新手友好度 72/100
githubnext/gh-aw-cao#16895 ·
维护者通常 1 天内回复
-
`helios / deploy`: switch zone wait in `deploy.sh` has almost no headroom over healthy startup times可能已有人在做 关联的 PR 仍在进行中或已合并。 未关闭Test Flake
难度 2/5 1-3 小时 新手友好度 74/100
oxidecomputer/omicron#11453 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 65/100
-
data only (no code) enhancement good first issue
难度 2/5 1-3 小时 新手友好度 88/100
维护者通常 1 天内回复