菜单
Run the complete AI4Math paper-to-skill workflow from a user-provided PDF or Markdown paper. Use when a coding agent should preserve the original paper, convert PDF to paper.md when needed, extract proof-pattern SkillCandidates and SkillCards with line evidence, and optionally route reviewed cards into cross-paper synthesis.
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。
基于 SOC 职业分类
Route AI4Math paper-reading tasks to the appropriate bundled skill package.
Use when extracting reusable AI4Math research skills from papers, especially PDF-to-SkillCard workflows, paper triage, PDF preparation, proof-pattern extraction, and cross-paper SkillCard synthesis for a coding-agent-neutral paper-to-skill workflow.
Cluster, deduplicate, generalize, and rank proof-pattern SkillCandidates or SkillCards extracted from multiple papers. Use when a coding agent should synthesize single-paper extractions into a domain method map, merged SkillCards, and a synthesis report for an AI4Math skill library.
Search for papers that may provide new methods, new theory, sharper bounds, different assumptions, proof techniques, lower bounds, abstractions, or cross-domain transfers for a math research direction. Use when a coding agent should look beyond nearest-neighbor related work and generate innovation_candidates.json with reasons and risks.
Download open-access research-paper PDFs selected from reading_plan.json, candidate_papers.json, or innovation_candidates.json. Use when a coding agent should save paper.pdf, metadata.json, and a bilingual download report for confirmed paper IDs without bypassing paywalls, and before PDF-to-Markdown conversion or paper-to-skill extraction.
Extract reusable mathematical research skills from Markdown papers, especially proof-pattern skills such as descent lemmas, inequality relaxations, telescoping bounds, error decompositions, Lyapunov arguments, concentration bounds, Galerkin/Cea arguments, Aubin-Nitsche duality, and Lax-Milgram well-posedness. Use when a coding agent is asked to turn a math paper, proof, theorem section, convergence analysis, stability analysis, or error analysis into SkillCandidate JSON, SkillCard YAML, or a paper-to-skill report without relying on external LLM APIs.
| name | paper-to-skill-workflow |
| description | Run the complete AI4Math paper-to-skill workflow from a user-provided PDF or Markdown paper. Use when a coding agent should preserve the original paper, convert PDF to paper.md when needed, extract proof-pattern SkillCandidates and SkillCards with line evidence, and optionally route reviewed cards into cross-paper synthesis. |
Use this skill as the default user-facing entrypoint. A user should be able to provide a paper.pdf and an output directory, then receive reusable proof_pattern SkillCandidates, SkillCards, and a review report.
If the user already provides paper.md, skip conversion and run extraction directly. Still keep the workflow-level artifact structure.
Accept:
paper.pdf as the primary user input.paper.md when the paper has already been converted.metadata.json.human_feedback_state.json.outputs/<research_project_id>/papers/<paper_id>/.paper.pdf
-> pdf-to-markdown-converter
-> paper.md
-> paper-to-skill-extractor
-> skill_candidates.json + skill_cards/*.yaml + report.md
For already-converted papers:
paper.md
-> paper-to-skill-extractor
-> skill_candidates.json + skill_cards/*.yaml + report.md
outputs/<research_project_id>/papers/<paper_id>/ when the user is running a project, otherwise use a clear local output folder derived from the input filename.paper.pdf when a PDF is provided.pdf-to-markdown-converter to create paper.md, mineru/, and conversion_report.json.paper.md as the source for line-referenced extraction evidence.human_feedback_state.json exists, pass it through to the extraction step.paper-to-skill-extractor to produce skill_candidates.json, skill_cards/*.yaml, and report.md.cross-paper-skill-synthesizer.Before finishing, confirm:
paper.pdf is preserved when provided.paper.md exists.skill_candidates.json is valid JSON.source.paper_md, source.start_line, and source.end_line.report.md separates accepted, needs-review, and rejected candidates.