dev-builder
Skill19 starsChanged 29 days agoMITView on GitHub
Implement a defined product or maintenance change, integrate results and verify relevant behavior. Use bug-fixer for unknown failure causes; independent review depends on concrete risk.
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- bug-fixerSkill
- design-brief-builderSkill
- design-makerSkill
- dev-plannerSkill
- goal-writerSkill
- product-spec-builderSkill
- project-knowledgeSkill
- release-builderSkill
- reviewerSkill
- self-evolverSkill
- skill-builderSkill
- creatorSkill
The file
Files
开发实施
Purpose
完成最小完整变更,证明可观察行为并留下下一会话所需的当前事实。
Trigger
目标、预期行为、范围和验收足够明确;可以是产品短任务、Phase 或非诊断型维护。
Required context
现状入口、有效需求/设计、相关任务或 Phase、工作区修改、验证入口和授权;只读相关材料。
Workflow
- 确认路线、规模和具体影响。符合 Phase 条件却缺计划时先补计划,其余短任务直接执行。
- 核对代码与他人修改,按当前有效契约实施;新决定替代旧规则时先对账,不以代码差异静默改需求。
- 用针对性行为检查验证;必要构建/测试通过后,有新变更、失败或疑点才重复。
- 验证后更新受影响现状、需求状态、设计和计划;无影响的入口不机械刷新。
- 按任务可分离性与收益选择有边界的子代理,明确文件归属并由主会话集成;风险决定独立审查而非是否可委派。R0/R1 不自动审查;轻量维护进入 batch.md,小型产品变更保留需求/短任务。R2 使用 fresh reviewer,尊重用户/宿主限制。
- 聚焦处理审查发现,有实质修复可再复核,无进展或资源预算用尽时保存恢复检查点。
- 在安全收尾点使用 project-knowledge 沉淀明确项目准则和受影响现状;撤销能力时关闭失效任务及测试义务,不能把历史未验收项重新列为待办。完成整个已授权目标和必要集成后交接,不因一个子任务结束提前退出。
Output
实现结果、相关契约与文件、验证结果及适用基线、更新的文档、按需拆分的实施/验证/审查/发布状态和下一步。
Stop or escalate
未知正确行为影响方案时只暂停相关部分。不能以验证缺口代替完成,也不能因为工具失败或多个文件就自动升 R2。外部动作按已有明确授权边界执行。
References
交付时读 stage-contract.md;文档、队列及审查细则读共同契约。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

