来源信息
- 仓库
- decodingai-magazine/building-a-coding-agent-from-scratch-course
- 最近来源活动
- 2026年6月26日 08:31
- 检测到的 SKILL.md 语言
- 英语
- 星标
- 314
- 分支
- 82
安装方式
默认使用会先检查来源的 Prompt;你也可以切换为直接命令,或下载本地副本。
检查来源文件
决定是否安装前,请先阅读 SKILL.md,以及 SkillsMP 当前展示的配套文件。
菜单
默认使用会先检查来源的 Prompt;你也可以切换为直接命令,或下载本地副本。
决定是否安装前,请先阅读 SKILL.md,以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/decodingai-magazine/building-a-coding-agent-from-scratch-course --skill commit命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
Demo skill that web-fetches two Decoding AI knowledge-graph articles, has the agent itself distill them into a typed entity/relation graph, and renders an interactive dark-themed force-directed KG into one self-contained kg.html — no graph library, no CDN.
Manual end-to-end QA playbook for decode against a real provider — what to type at each surface (chat, gated read/write, bash, todo, web_fetch, ask_user, lsp, agent subagents, docker/modal sandboxing, headless `decode run`, HITL, model override, replay, mid-turn steer/abort, persistence/memory, /ship, sandbox git token) and what "working" looks like. Use when manually testing or verifying a decode surface end-to-end.
Demo skill for the meta "decode improves decode" flow — launch decode in a sandboxed clone of the course repo, plan and build one small feature inside the Workspace, let Hand-back push the Session Branch, then open a draft PR against the course repo.
基于 SOC 职业分类
正在显示 SKILL.md
| name | commit |
| description | Stage the appropriate changes and commit them with a Conventional Commits message. |
You commit the work in the current working tree autonomously: you stage the right files, write
the message, and run git commit. You commit exactly what you stage.
git status and git diff (and git diff --cached for anything already
staged). If there is nothing to commit, say so and stop.git add. Do not stage unrelated edits, build
artifacts, or secrets; if the tree mixes unrelated changes, stage the coherent subset and say what
you left out.type(scope): summary — type ∈ feat, fix, refactor, docs, test, chore, build, ci,
perf; imperative summary ≤ 72 chars.git commit with that message, committing exactly what you staged. Describe only changes you
actually saw in the diff; never invent them. Report the resulting commit subject.