[curriculum-eval] side-quest-17-07-repo-poisoning.md: cognitive_load — 1,234 words and concept overload in a security side quest
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 2/5
- 预计耗时
- 1-3 小时
- 新手友好度
- 82/100
- Issue 类型
- 文档
- 描述清晰度
- 描述清楚
- 活跃度
- 活跃
- 技术栈
- github, github-actions, markdown
- 领域
- content, documentation
调研方向
阅读 workshop/side-quest-17-07-repo-poisoning.md 和仓库的 AGENTS.md 规范。在链接所指向的网站上核对 GitHub Agentic Workflows 文档链接，然后修订指定章节，将页面精简到约 900 词，同时保留目标、两个练习和检查点。完成标准是三种防御措施各有一个以验证操作结尾的简短小节；运行指定的 markdownlint 命令。
由索引模型根据 Issue 内容生成。
描述
File: workshop/side-quest-17-07-repo-poisoning.md
Overall Score: 5.0 / 10.0 (corpus mean: 6.1; lowest in Part 2)
Flagged Dimensions:
| Dimension | Score | Benchmark | Delta |
|---|---|---|---|
| cognitive_load | 5.8 | ≥ 9 | −3.2 (1,234 words; concept_overload) |
| active_learning | 2.7 | 10 (density ≥ 3) | −7.3 |
| checkpoint_quality | 0.0 (rubric false negative, see rubric issue) | 10 | n/a |
Root Cause (≤ 2 sentences):
The page teaches the attack, why it matters, three defences (contents: read, safe-outputs: create-pull-request, network.allowed-domains), and two exercises in one sitting; the "How AW Defends Against It" section alone spans lines 45–124. Concept count exceeds the 15-concept ideal while exercises are small relative to prose.
Evidence (quoted from the file):
"Apply the three gh-aw defences:
contents: read,safe-outputs: create-pull-request, andnetwork.allowed-domains."
Learning Science Rationale:
Cognitive Load Theory (Sweller) holds that novices cannot hold many interacting elements at once; presenting each defence followed immediately by a small apply task (worked-example then completion problem) lowers intrinsic load and moves learners from "understand" toward "apply/analyse".
Improvement Prompt (for an agent):
Edit workshop/side-quest-17-07-repo-poisoning.md to bring it under ~900 words: trim the "Why This Matters" and "What You Can Do as a Workflow Author" prose, and restructure "How AW Defends Against It" into three short subsections, each ending with a one-line verification action (a code block or checklist item). Keep the learning objectives, both exercises and the :white_check_mark: Checkpoint. Follow AGENTS.md conventions, verify links against https://github.github.com/gh-aw/, and run `npx --yes markdownlint-cli2 "**/*.md"`.
Expected Score After Fix: 6.8 / 10.0
Generated by 🔬 Curriculum Quality Evaluator · copilot · auto · ~35 AIC · ⌖ ~1 AIC · ⊞ 5K · ◷
- expires on Nov 13, 2026, 3:32 AM UTC
- 主要语言
- JavaScript
- 星标
- 62
- 派生
- 43
- 平均合并
- 1 天 5 小时
- 30 天内合并 PR
- 9
环境准备
在浏览器里用你自己的 GitHub 账号启动这个项目的开发容器。
- 没有 Dockerfile 或 Docker Compose 文件
- 没有 Pull Request 模板
- 没有贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
githubnext/gh-aw-workshop 的其他 Issue
-
dedup documentation
难度 2/5 1-3 小时 新手友好度 82/100
githubnext/gh-aw-workshop#5628 ·
维护者通常 1 天内回复
-
dedup documentation
难度 2/5 1-3 小时 新手友好度 68/100
githubnext/gh-aw-workshop#5629 ·
维护者通常 1 天内回复
-
dedup documentation
难度 2/5 1-3 小时 新手友好度 84/100
githubnext/gh-aw-workshop#5627 ·
维护者通常 1 天内回复
-
curriculum documentation quality
难度 2/5 1-3 小时 新手友好度 78/100
githubnext/gh-aw-workshop#5625 ·
维护者通常 1 天内回复
-
curriculum documentation quality
难度 2/5 1-3 小时 新手友好度 84/100
githubnext/gh-aw-workshop#5622 ·
维护者通常 1 天内回复
查看 githubnext/gh-aw-workshop 的全部 Issue
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 78/100
-
good first issue help wanted javascript up-for-grabs
难度 2/5 1-3 小时 新手友好度 78/100
andrewsiah/agentswarm-data-tasks#307 ·
维护者通常 1 天内回复
-
needs-triage
难度 2/5 1-3 小时 新手友好度 78/100
coder/claudecode.nvim#588 ·
维护者通常 1 天内回复
-
area: web tech debt
难度 2/5 1-3 小时 新手友好度 82/100
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 82/100
digitalbazaar/http-client#60 ·