CCDawn/briefbound-skills / skills/engineering/briefbound-router
briefbound-router
VerifiedUse 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.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 10.8 KB |
| agents/openai.yaml | config | 375 B |
| references/capability-routing.md | markdown | 5.2 KB |
| references/collaboration-discovery.md | markdown | 1.7 KB |
| references/evaluation.md | markdown | 989 B |
| references/github-skill-candidates.md | markdown | 11.0 KB |
| references/harness-compat.md | markdown | 3.6 KB |
| references/output-forms.md | markdown | 2.9 KB |
| references/review-test.md | markdown | 2.6 KB |
| references/routing-practice.md | markdown | 9.8 KB |
| references/runtime.md | markdown | 3.2 KB |
| references/ui-preview-approval.md | markdown | 3.3 KB |
references/collaboration-discovery.md
View on GitHub# Collaboration Discovery
只在需求已收敛、功能 owner 已选定、本轮原生 thread/list 能力可用,且现有同项目独立会话可能互助时读取。能力不可用时返回 `NONE`,不得伪造或创建候选。发现目的是判断平级协作价值,不是启动协调。
## 有界发现
1. 使用本轮原生 thread/list 能力查找同一项目的活跃对话;registry 已存在时读取 Agent、scope、branch/worktree 和 checkpoint,不为发现初始化 memory 或 registry。
2. 只读最多 3 个最相关候选,确认真实任务、进度、有效 thread id、互补价值和重叠风险;路径或名称相似不等于同项目。
3. 比较对各自任务的帮助、重复工作消除、冲突/返工降低与消息/协调/合并成本。候选忙碌、过期、共享同一小文件或协作会拖慢任一关键任务时排除。
## 决策
- `NONE`:无正收益候选,原 owner 静默继续。
- `PEER_CONTEXT_REVIEW`:同行上下文能改善当前判断,交 thread coordination 做一次只读建议。
- `PEER_READ_ONLY`:双方交换独立只读证据能帮助各自任务,交 thread coordination 提议协作。
- `COORDINATE_OVERLAP`:只有重叠、讨论或恢复需求,交 thread coordination。
- `PEER_DISJOINT_WRITE -> PEER_COLLABORATION_READY`:至少两个现有会话各有原任务,存在明确依赖或共同集成面,写入可分离且整体收益高于协商成本,交 multi-agent orchestration。
- `ASK_CREATE`:无现有候选但新增会话有明确收益;向用户询问一次,未授权则当前 owner 继续。
发现时不得发送消息、claim、创建会话或暂停其他 Agent。不得把现有会话视为可派发的 worker;路由后只能提出平级协作。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 472k | Updated | Claude CodeCodex | Verified | Install |
| 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: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| 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 available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |