Community생산성 & 협업github.com

setup-roadmap

Shows an agent how to plan or hand off "setup roadmap" so the concrete output is usable copy or a script. The procedure moves through 实验装置与技术路线图 Skill, 输入约定, 工作流程(7 步,按序执行). Use it when a new assignment calls for this craft.

setup-roadmap란 무엇인가요?

setup-roadmap is a Claude Code agent skill that shows an agent how to plan or hand off "setup roadmap" so the concrete output is usable copy or a script. The procedure moves through 实验装置与技术路线图 Skill, 输入约定, 工作流程(7 步,按序执行). Use it when a new assignment calls for this craft.

지원 대상~Claude Code~Codex CLI~Cursor
npx skills add https://github.com/qingyulzx/pku-efficiency3-skills/tree/HEAD/apparatus-roadmap

Installed? Explore more 생산성 & 협업 skills: steipete/gemini, steipete/gh-issues, steipete/skill-creator · View all 6 →

즐겨 사용하는 AI에게 물어보기

이 에이전트 스킬이 미리 로드된 새 채팅을 엽니다.

문서

实验装置与技术路线图 Skill

把「装置怎么搭、步骤怎么走、物料从哪来到哪去」画成一张对得起材料的图。 判断(部件是什么类型、箭头朝哪、阶段怎么分)由你做;解析、渲染、校验交给脚本。 核心承诺:图上的每个元素都能逐字回溯到输入材料,材料提到的每个实体要么画了、 要么有书面的省略理由——这是由 check_consistency.py 机器强制的,不是自评。

输入约定

  • /input/case.json:任务说明(含 output_contract 字段名,如有)。
  • 材料文件(按 case 指定):PDF 用容器 pdf-text-tool 取文字层,扫描页用 ocr-tool;两者都失败的页记入 uncertainty,不猜。
  • 参考 /skills/weekly-skill/references/:ir_schema.md(IR 格式)、symbol_library.md(图元/线型约定)、 extraction_playbook.md(判断规则)、output_contract.md(result.json 字段)。

工作流程(7 步,按序执行)

Step 1 读取与盘点

读 case.json 和全部材料。PDF 只信真解析结果。列出材料清单(文件、页数、 可解析性)——这份清单后面写进 provenance。

Step 2 保底产出

立即初始化 result.json(所有契约字段 null + 原因):

python /skills/weekly-skill/scripts/finalize_result.py --init --case /input/case.json --out /app/output/result.json

If that command fails for any reason (wrong CWD, import error, anything), do NOT proceed without the floor: write /app/output/result.json by hand right now — case_id plus every output_contract field as explicit null with a reason, status "partial". The skeleton must exist before any other work. This skill is installed at /skills/weekly-skill; always invoke its scripts by absolute path.

此后每完成一步,回头把对应字段从 null 换成实值。中途崩溃也留下完整骨架。

Step 3 抽取与判断(agent-in-the-loop)

按 extraction_playbook.md 规则,亲自完成:

  1. 部件识别:每个部件 → id/label/type(映射见 symbol_library.md)/specs;
  2. 连接拓扑:from/to(方向对材料里的动词确认)、medium、kind、conditions;
  3. route 类图:节点 kind(stage/step/decision/start/end)、step_no 与正文对照、 回环边标 kind=loop + route=below;
  4. 每个元素挂 source {location, quote}(quote 逐字复制);规格未给 → certainty=unspecified_spec;连接靠推断 → inferred;
  5. extraction_index 登记所有提到的实体/步骤(mapped_id 或 omitted+reason)。

产出一个 IR JSON(格式照 ir_schema.md),存 /app/output/intermediate.json。

Step 4 IR 校验

python /skills/weekly-skill/scripts/validate_ir.py /app/output/intermediate.json

exit 0 才继续。常见错误:id 重复、连接端点不存在、元素缺 source.location、 extraction_index 有 omitted 无 reason。改 IR 再跑。

Step 5 渲染

python /skills/weekly-skill/scripts/render.py /app/output/intermediate.json --outdir /app/output \
       --name figure --also-png --also-pdf --png-scale 2 \
       [--font <font-discovery 找到的 CJK 字体路径>]

产出 figure.svg(四层可编辑)+ figure.png + figure.pdf + figure.geometry.json。 检查 stdout:overflow_warnings 应为空;font_notes 里有「no CJK font」就用 font-discovery 工具找字体重跑;PNG 因字体缺中文时 SVG 仍是真文字(保真源), 同时在 result.json 的 uncertainty 里如实记录。

Step 6 双向一致性校验(不过不算完成)

python /skills/weekly-skill/scripts/check_consistency.py /app/output/intermediate.json \
       <材料文本文件> /app/output/figure.svg /app/output/figure.geometry.json \
       --out /app/output/consistency_report.json

五项检查:引文逐字回溯(防无中生有)、覆盖索引完备(防漏画)、SVG↔IR id 双向 一致、箭头落点无悬空、图层齐全。exit 1 必须回到 Step 3 修 IR 再重跑 5–6, 不许带病交付。材料文本 = 解析出的全文(多文件可先合并)。

Step 7 终化

把实值填进 result.json(字段与填法见 output_contract.md:图路径、元素来源表、 数据映射、provenance、uncertainty 与图上标记一一对应),然后:

python /skills/weekly-skill/scripts/finalize_result.py --finalize --result /app/output/result.json \
       --case /input/case.json --artifact-dir /app/output \
       --consistency /app/output/consistency_report.json

exit 0 才算交付。partial 结果必须在 run_status 里列明未完成步骤与影响。

编辑场景(已有图改部件/加管路)

改 IR,不改 SVG:

  1. 用户给的新材料存成文本文件;
  2. 写 edits 文件(操作格式见 scripts/edit_ir.py 文档头): add_component / add_connection / update_component / remove_component …;
  3. python /skills/weekly-skill/scripts/edit_ir.py intermediate.json edits.json --material 新材料.txt --out intermediate_v2.json——新增元素的 quote 会被逐字核验,核验不过整批拒绝;
  4. 对 v2 重跑 Step 4→6,result.json 的 provenance.change_records 自动带出。

异常处置路径

异常处置
部件规格未给照画 + certainty=unspecified_spec(图上自动「△ 规格未提供·示意」),不编数值
连接关系材料未写inferred 灰虚线标注,或不连线进未连线区;uncertainty 里说明
流程顺序歧义选一种读法,uncertainty 记录歧义原文与理由
文字溢出画布自动扩展;geometry.json 的 overflow_warnings 非空时缩小标签文本重渲
字体缺失(中文变框)font-discovery 找字体传 --font;仍无 → SVG 保真文字 + uncertainty 记录
导出失败SVG 是保底交付件;PNG/PDF 失败写进 font_notes/uncertainty,不谎报
材料 PDF 不可解析该部分字段 null + 原因(页码),不猜
校验硬失败修 IR 重跑,绝不跳过或降级检查

自查清单(交付前)

  • result.json 全字段非 null 或 null+原因(finalize exit 0)
  • consistency_report.json pass=true
  • SVG 打开可见四层(frame/structure/annotation/source),元素带 data-id
  • PNG/PDF 存在且非空;字体问题(如有)已记录
  • 图上每个 △/灰虚线在 uncertainty 里有一一对应条目
  • intermediate.json 已随交付保存(改图真源)

限制

  • 示意图元不表达工程比例与真实尺寸;不判断实验方案科学合理性。
  • 中文 PDF 的 OCR 质量受原件影响,引文以文字层优先。
  • PDF 矢量导出依赖 cairosvg(可选);缺失时退化为位图 PDF 并如实标注。
  • 依赖:Python 3 标准库(SVG);Pillow(PNG/位图 PDF,可选)。

Individual skills in this repo

This repo contains 5 individual skills — each has its own dedicated page.

biopathway-mechanism-diagram

Build or edit publication-grade biological mechanism/pathway diagrams from research materials (PDF/text) as a challengeable mechanism statement - three-layer model (typed entities, typed directed edges with verbatim evidence, compartment layout), semantics-correct SVG source, PNG/PDF exports and a traceable, uncertainty-aware result.json. Use for tasks asking to draw, generate or edit signaling pathways, mechanism figures, crosstalk diagrams or pathway flowcharts; refuses to guess regulation direction and marks unstated information instead.

graphical-abstract-builder

Build a single-page graphical abstract (question, method, results, conclusion) from research material as a traceable, editable SVG with PNG/PDF exports. Use when asked to create or edit a graphical abstract, science summary figure, mechanism or flow diagram, or data chart from a paper, study notes or experimental records, especially when editable source files and element-to-source traceability are required.

paper-to-academic-ppt

Turn a research paper (PDF) into a structured, evidence-traceable, template-compliant academic slide deck. Two phases: agent-authored narrative spine and page-level claims (each with paper locators and verbatim quotes), then scripted template adaptation that writes content into real master/layout placeholders, embeds original paper figures, checks overflow/aspect/notes, and exports a PDF preview. Handles missing conclusions, figure-trimming, layout mismatch and over-limit slide counts with explicit rules.

relationship-schematic

Generate or edit scientific entity-relationship schematics (mechanism diagrams, pathway chains, experimental workflows, structural containment) as traceable, editable SVG with verified relation semantics. Extracts an evidence-quoted entity/relation table first, renders a layered SVG from it, reads the drawing back to verify every relation type, direction and containment, then exports 300-dpi PNG and vector PDF. Use for any "draw a mechanism/workflow/relationship figure from this paper or case material" task where scientific correctness of entities, arrow directions and hierarchy matters.

slides-genedit

Generate and edit academic presentation decks from research material. Two routes: (1) generate a new deck from a paper/notes under fixed audience, duration and slide-count constraints, with per-slide claims, verbatim-number evidence mapping, speaker notes, and a rendered PDF preview; (2) source-preserving edits of an existing starting deck - inventory it, apply only the requested local changes, and prove untouched slides stayed byte-identical. Use when asked to turn research material into slides, or to modify/compress/re-audience an existing deck without breaking it.

관련 스킬