依据WCAG 2.2设计和检查Web、iOS及Android界面,核查读屏标签、键盘操作、配色对比度和操作目标尺寸,使界面支持辅助技术。
- 为纯图标按钮添加说明
- 检查键盘操作与模态框焦点
- 检查对比度与操作目标尺寸
将大型功能开发或迁移目标拆成可分多次执行的计划,整理每个步骤的背景、依赖关系和验证方法,支持将工作交接给其他代理。
Turn a one-line objective into a step-by-step construction plan for multi-session, multi-agent engineering projects: one-PR-sized steps with self-contained context briefs, dependency graph with parallel-step detection, adversarial review gate, and plan mutation protocol. Use when planning a large feature, refactor, or roadmap that spans multiple PRs or sessions; not for single-PR tasks or when the user says "just do it".
blueprint将大型开发目标拆成以PR(提交审核的一组更改)为单位的执行计划。它检查项目结构和现有计划,梳理依赖关系、可并行执行的步骤及回退方法。保存在plans/中的Markdown文件包含每步的背景、任务、验证命令和完成条件,让其他代理也能据此开始工作。
适用于跨多个PR或多次会话的功能开发、重构和迁移。每个步骤都集中列出所需信息,有助于减少人员交接或会话间信息缺失造成的返工。计划由另一个子代理进行批判性审查,修正重大问题后再定稿。
使用/blueprint需要Claude Code。有Git和GitHub CLI时,计划会包含分支、PR和CI流程;没有时则切换为直接编辑方式。不适合单个PR即可完成的小任务,也不适合要求立即开始实现的请求。
本介绍由 AI(OpenAI Codex)根据公开的 SKILL.md 撰写。准确的规格请查看下方原文。
AI 仅根据 SKILL.md 的内容,从 4 个方面进行了评估。星标数和热度不计入评估。
总分按每项 A=25 分、B=15 分、C=5 分相加。 由 AI(OpenAI Codex)于 2026年10月19日 评估。仅供参考,安装前请同时查看原文。
安装前可以查看交给 AI 代理的指令内容。
Turn a one-line objective into a step-by-step construction plan that any coding agent can execute cold.
Do not use for tasks completable in a single PR, fewer than 3 tool calls, or when the user says "just do it."
Blueprint runs a 5-phase pipeline:
plans/. Every step includes a context brief, task list, verification commands, and exit criteria — so a fresh agent can execute any step without reading prior steps.Blueprint detects git/gh availability automatically. With git + GitHub CLI, it generates full branch/PR/CI workflow plans. Without them, it switches to direct mode (edit-in-place, no branches).
/blueprint myapp "migrate database to PostgreSQL"
Produces plans/myapp-migrate-database-to-postgresql.md with steps like:
/blueprint chatbot "extract LLM providers into a plugin system"
Produces a plan with parallel steps where possible (e.g., "implement Anthropic plugin" and "implement OpenAI plugin" run in parallel after the plugin interface step is done), model tier assignments (strongest for the interface design step, default for implementation), and invariants verified after every step (e.g., "all existing tests pass", "no provider imports in core").
.md files — no hooks, no shell scripts, no executable code, no package.json, no build step. Nothing runs on install or invocation beyond Claude Code's native Markdown skill loader.This skill ships with ECC. No separate installation is needed when ECC is installed.
If you are working from the ECC repository checkout, verify the skill is present with:
test -f skills/blueprint/SKILL.md
To update later, review the ECC diff before updating:
cd /path/to/ECC
git fetch origin main
git log --oneline HEAD..origin/main # review new commits before updating
git checkout <reviewed-full-sha> # pin to a specific reviewed commit
If you are vendoring only this skill outside the full ECC install, copy the reviewed file from the ECC repository into ~/.claude/skills/blueprint/SKILL.md. Vendored copies do not have a git remote, so update them by re-copying the file from a reviewed ECC commit rather than running git pull.
/blueprint slash command)Inspired by antbotlab/blueprint — upstream project and reference design.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
依据WCAG 2.2设计和检查Web、iOS及Android界面,核查读屏标签、键盘操作、配色对比度和操作目标尺寸,使界面支持辅助技术。
从指令、记忆、工具执行和界面显示等十二个层面排查人工智能代理故障,依据代码和日志梳理原因,按严重程度列出问题并提出修复方案。
用实际开发任务比较多个编程智能体,记录成功率、可获取的API费用、耗时和重复运行的稳定性,为工具选型及更新后的性能评估提供数据。
设计和调整AI智能体使用的工具、输入与返回格式,明确出错后的恢复、重试和停止条件,管理上下文信息量,并以完成率、重试次数和成功任务成本评估改进。
根据错误信息和执行环境,分析AI代理反复失败或重复操作的原因。通过小范围、可撤销的恢复操作检验假设,并将结果与依据整理成可供交接的报告。
在AI代理反复失败或重复操作时,记录错误与环境前提、排查原因,采取影响有限的恢复操作,并汇总结果和证据,为有依据的重试与交接提供支持。