Back to Marketplace
30-day free campaign

Run this helper free — no credit card

Every helper is free for 30 days. Answer 3 questions and get the full result in 2 minutes.

Start free →
FREE
Scanned
Grow Business

Paper Survey Builder

围绕某个研究主题搭建可长期维护的 Obsidian 调研工作区。适用于主题仍然模糊、需要先重构研究问题与关键词,再由 Codex/subagent 扩论文池、归纳主脉络、生成逐文精读、补方法图与可比组说明,最后由脚本装配 preview 并做 Obsidian 校验。

👁 1 views · 📦 0 installs

Install in one line

mfkvault install paper-survey-builder

Requires the MFKVault CLI. Prefer MCP?

New skill
No reviews yet
New skill
💻 Codex
This helper was discovered by MFKVault crawlers from public sources. Original author retains all rights. To request removal: [email protected]
Community helper
This helper was discovered by MFKVault crawlers from public sources. MFKVault does not create, maintain, or guarantee the output of this helper. Results are AI-generated and may be incomplete, inaccurate, or outdated. Use at your own risk. Original author retains all rights. Request removal
FREE

Free to install — no account needed

Copy the command below and paste into your agent.

Instant access • No coding needed • No account needed

What you get in 5 minutes

  • Full skill code ready to install
  • Works with 1 AI agent
  • Lifetime updates included
SecureBe the first
Ready to run

Run this helper

Answer a few questions and let this helper do the work.

Advanced: use with your AI agent

Description

--- name: paper-survey-builder description: 围绕某个研究主题搭建可长期维护的 Obsidian 调研工作区。适用于主题仍然模糊、需要先重构研究问题与关键词,再由 Codex/subagent 扩论文池、归纳主脉络、生成逐文精读、补方法图与可比组说明,最后由脚本装配 preview 并做 Obsidian 校验。 --- # Paper Survey Builder ## 概览 - 这是一个 `agent-first` skill:先由 Codex 明确范围、拆任务、并行组织 subagent,再让脚本处理结构化落盘、图候选提取、preview 装配与校验。 - 目标不是“尽快吐出一篇固定格式报告”,而是沉淀一套可回溯、可复查、可长期维护的调研工作区。 - 默认把 `index-style` 当作深度调研主模式;`master-style` 保留为轻量长期维护视图。 ## 角色分工 ### Codex / Subagent 负责 - 模糊需求重构为明确调研问题 - 关键词簇、排除项、时间范围、venue 范围、近邻方向拆分 - 扩论文池、核 venue/year/repo/OpenReview/arXiv - 流派/主脉络归纳与边界说明 - 单篇 note 的非结构化理解与写作 - benchmark 可比性判断、不可横比说明 - 方法图语义核对与 Mermaid 兜底图生成 ### 脚本负责 - 初始化 staging 工作区与 schema - seed/PDF 规范化为 `candidate_inventory.csv` - 外部调研结果合并回 staging - 提取图候选并生成 `figure_manifest.csv` - 装配 preview、复制图文件、维护相对路径 - Obsidian lint、图文一致性基础检查、dry-run publish ## 推荐流程 1. `scope` Codex 根据用户原始需求生成 `scope_brief`、`query_map`、范围边界与排除项。 2. `discover` `paper-scout` 等 subagent 并行扩论文池、核基础事实,并把结果沉淀到 `sources/`。 3. `inventory` 脚本把 seed/PDF/外部结果规范化成 `candidate_inventory.csv`;经人工或 subagent 核验后,回写 `validated_inventory.csv`。 4. `lineage` `lineage-architect` 生成 `lineage_map.json` 与 `paper_assignments.csv`。 5. `paper-briefs` `paper-summarizer` 为每篇论文输出语义 brief,放到 `drafts/paper_briefs/`。 6. `benchmarks` `benchmark-normalizer` 补 `benchmark_records.csv` 与不可横比说明。 7. `figures` 脚本先提图,`figure-verifier` 再做语义核对;若没有可信方法图,再生成 Mermaid fallback。 8. `assemble` 脚本读取结构化表、paper brief、report blocks、figure manifest,生成 `preview/`。 9. `validate / publish` 校验链接、embed、图路径、证据锚点与图文一致性;默认先 dry-run,再决定是否发布。 ## 默认入口 新的薄工具层 CLI 只暴露结构化动作: ```powershell python scripts/survey_pipeline.py init-workspace --topic "主题名" --vault "Obsidian Vault" --target-dir "主题目录" --mode index python scripts/survey_pipeline.py build-candidate-inventory --staging-root "<staging_root>" --seed-file "papers.csv" python scripts/survey_pipeline.py merge-research-results --staging-root "<staging_root>" --validated-inventory "validated_inventory.csv" --paper-briefs-dir "paper_briefs" python scripts/survey_pipeline.py extract-figure-candidates --staging-root "<staging_root>" python scripts/survey_pipeline.py render-preview --staging-root "<staging_root>" --mode index python scripts/survey_pipeline.py validate-preview --staging-root "<staging_root>" python scripts/survey_pipeline.py publish-preview --staging-root "<staging_root>" --target-dir "主题目录" --apply ``` ## 结构化产物 - `scope/` - `scope_brief.md` - `scope_brief.json` - `query_map.json` - `data/` - `candidate_inventory.csv` - `validated_inventory.csv` - `lineage_map.json` - `paper_assignments.csv` - `benchmark_records.csv` - `figure_manifest.csv` - `drafts/paper_briefs/` 每篇论文一个语义 brief,至少包含: - `scientific_problem_1_7` - `task_dataset_summary` - `baseline_summary` - `result_summary` - `remarks` - `uncertainty_notes` - `evidence_anchors` - 可选 `figure_mermaid` - `drafts/report_blocks/` 用于主文档总述、脉络说明、总观察、前沿扩展和附录说明。 - `logs/` 保留 validate / publish 报告,并单独记录 `lineage_review.md`、`figure_review.md`。 ## 方法论约束 - 没有 `scope_brief`,就不要开始大量搜文献。 - 没有 `validated_inventory.csv`,就不要写正式主索引和结论。 - 没有 `paper_brief` 与 `evidence_anchors`,就不要让脚本代写逐文精读内容。 - 不能横向比较时,必须显式写出 `不可直接横比` 或等价说明。 - 方法图优先级固定为 `arxiv-source > pdf-figure > pdf-extraction > page-crop > mermaid-fallback`。 - Mermaid 只作为方法图兜底,不替代真实可信的原始图。 - staging 是工作区,不要直接在正式 vault 里边搜边改。 ## 能力复用 - 文献发现优先参考 `research-lookup`,结果保存回 `sources/` - 方法图提取优先复用 `extract-paper-images` - 本地复用与去重遵循 `paper-search` 风格 - Mermaid fallback 建议参考 `markdown-mermaid-writing` - 图像路径修复与语义核对可借鉴 `obsidian-paper-figure-repair` - 统计图仍优先 `seaborn`,需要细控制时再用 `matplotlib` ## 并行与人工关卡 - `scope` 完成前不并行拆任务。 - `discover`、`paper-briefs`、`benchmark normalization`、`figure verification` 可以并行。 - `lineage` 定稿、最终结论收口、assemble、publish 必须串行。 - 同一篇论文的 `paper-summarizer` 与 `figure-verifier` 可以并行,但最终 note 合并必须串行。 - 即使强自动化,也保留两个必须的人审门: - 主脉络与口径锁定后,再大批量补条目。 - validate 通过并人工看过报告后,再 `publish --apply`。 ## 资源索引 - `references/workflow-overview.md`:agent-first 总流程、阶段门禁、失败回退 - `references/report-modes.md`:`index-style` 与 `master-style` 的职责边界 - `references/evidence-policy.md`:证据锚点、不确定性、外部采证与保存规则 - `references/figure-policy.md`:图源优先级、图文一致性、Mermaid fallback - `references/obsidian-style.md`:表格、链接、embed、assets 命名约束 - `references/visualization-recipes.md`:结构化图表 helper 的使用方式 - `references/subagent-playbook.md`:subagent 工种契约、输入输出与并行边界 - `references/quality-gates.md`:交付前检查清单 - `assets/templates/`:`index-style` 与 `master-style` 模板 - `assets/schemas/`:workspace 关键文件的 schema 样例 - `scripts/survey_pipeline.py`:薄工具层主入口 ## 输出要求 - 默认中文叙述,保留英文论文标题、venue、指标原文。 - 逐文精读的结构化字段由脚本装配,非结构化判断必须来自 Codex/subagent。 - 若证据不足,就明确写 `待补` 与缺口类型,不要用脚本或模板伪造结论。

Preview in:

Security Status

Scanned

Passed automated security checks

Time saved
How much time did this skill save you?

Related AI Tools

More Grow Business tools you might like

codex-collab

Free

Use when the user asks to invoke, delegate to, or collaborate with Codex on any task. Also use PROACTIVELY when an independent, non-Claude perspective from Codex would add value — second opinions on code, plans, architecture, or design decisions.

Run free

Rails Upgrade Analyzer

Free

Analyze Rails application upgrade path. Checks current version, finds latest release, fetches upgrade notes and diffs, then performs selective upgrade preserving local customizations.

Run free

Asta MCP — Academic Paper Search

Free

Domain expertise for Ai2 Asta MCP tools (Semantic Scholar corpus). Intent-to-tool routing, safe defaults, workflow patterns, and pitfall warnings for academic paper search, citation traversal, and author discovery.

Run free

Hand Drawn Diagrams

Free

Create hand-drawn Excalidraw diagrams, flows, explainers, wireframes, and page mockups. Default to monochrome sketch output; allow restrained color only for page mockups when the user explicitly wants webpage-like fidelity.

Run free

Move Code Quality Checker

Free

Analyzes Move language packages against the official Move Book Code Quality Checklist. Use this skill when reviewing Move code, checking Move 2024 Edition compliance, or analyzing Move packages for best practices. Activates automatically when working

Run free

Claude Memory Kit

Free

"Persistent memory system for Claude Code. Your agent remembers everything across sessions and projects. Two-layer architecture: hot cache (MEMORY.md) + knowledge wiki. Safety hooks prevent context loss. /close-day captures your day in one command. Z

Run free