Skip to content

CCDawn/briefbound-skills / skills/engineering/briefbound-router

briefbound-router

Verified

Use when Chinese-first active intent discovery, local context probing, user calibration before downstream skill selection, owner/workflow routing, continuation, permission inference, or multi-thread/conflict coordination is needed.

references/runtime.md

View on GitHub
# Briefbound Router 长任务与交接

用于明确开放目标、跨会话恢复、跨阶段交接或持久风险。普通有限任务由当前 owner 直接完成。

## 最小状态与续接

只保留目标、当前 owner/推进项、关键决定、最新证据、未决风险及下一动作。这是对话内状态,不自动创建账本或 dashboard。确需持久化且已有授权时由 project-memory 保存最小 delta。

用户已授权实现/修复/优化时连续推进;完成一个 task 不构成用户闸门。“继续/确认/按推荐”重验当前项后续接,不重开已解决的对齐。意图变化、范围扩大、不可逆动作、真实冲突或关键取舍才暂停;只报告新结果和会影响行动的风险。

## 实施与验证

只有可审阅方案或真实设计风险需要 planning;仅独立产物、owner、依赖或验收边界值得拆任务。同机制机械修改合并;测试与文档归入所服务的功能。

每项重验写入面与现有改动,实施最小结果,运行风险相称的验证。失败先定位实现、环境、测试意图或需求原因;契约内修复并重验,当前项通过后自动进入下一个已授权项。普通配置或实验不强制 TDD。Windows 上 `python`/`py` 命中 WindowsApps 别名时,先试内置运行时再扩大搜索。

完成必须满足需求、验证通过、无未披露风险且未越界。测试、运行时、打包、远程状态分别取证;实验比较固定 baseline/data/protocol。已知残留才清理;集成前可明确 DEFERRED_INTEGRATION,集成后不能仍用它收口。PR 准备度交 briefbound-pr-review。

## 开放目标循环

仅当用户明确要求持续/反复迭代、无法预拆有限步骤、每轮需要据证据选动作,且无研究/评分/协作专项 owner 时使用。普通“继续”不自动建立新循环。

继承目标、证据、约束和停止条件;选最小高信息动作,执行后据新证据推进或改方向,不每轮询问是否继续。目标达成、预算或权限边界、用户暂停、反复无新信息时停止并说明现场。一次失败不等于无路可走。定时继续或监控使用产品调度,不以文字承诺后台运行;循环不自动授予持久化或远程动作权限。

## 正式交接

从现有需求和产物提取:已完成结果与证据、未完成项、保护边界、未决风险触发条件和下一动作。合并重复信息,不重跑仍适用的昂贵检查;缺工作标部分完成,缺证据标未验证,不能用总结代替实现。

普通单 owner 任务直接简洁汇报。正式交接也不自动授权提交、推送、发布或补功能;需要保存时遵守已有持久化授权。

## 多会话与阻塞

代码/Git 是实现事实源,live registry 管协调状态;共享写入与恢复遵循 briefbound-thread-coordination。pause 需 PAUSED 回执;owner 失活由注册 Agent takeover 并继承恢复义务;resolve 后通知并恢复,resumePendingAgentIds 清零才闭环。

阻塞报告原因、受影响项、已尝试路径、保留的证据、恢复方式及不可约输入。难、慢或一次验证失败本身不构成阻塞;仍有授权内的有效动作就继续。
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall