tomevault-io Paper Analyst
Analyze academic papers from PDF or pasted text. Output in Chinese by default.
All outputs follow references/output-schema.md. Paper type detection uses
references/paper-type-rubric.md. Anti-hallucination rules in
references/quality-checklist.md.
Quick Reference
| File | Purpose |
|---|---|
references/output-schema.md |
Section structure and field rules |
references/paper-type-rubric.md |
How to classify paper type |
references/quality-checklist.md |
Anti-hallucination checklist |
references/presentation-schema.md |
Slide plan JSON schema |
references/presentation-style-guide.md |
Content compression rules for slides |
references/pptx-handoff.md |
How to call the pptx skill for rendering |
scripts/extract_pdf_meta.py |
Optional: extract PDF metadata to JSON |
Mode Selection
Default mode: standard. Detect from user's request:
| Mode | Trigger | Output |
|---|---|---|
quick |
"quick", "简单说", "一句话", "简要" | Header + info + abstract + 3 contributions |
standard |
(default) | Full analysis: sections 1–5 |
extended |
"前作", "课题组", "prior work" | standard + author/group prior work |
presentation |
"PPT", "组会", "汇报大纲", "slides" | standard + slide outline |
presentation_with_figures |
"图表", "figures", "带图", "关键图" | presentation + figure annotations |
If ambiguous, use standard and offer to switch.
Workflow
Step 1: Assess Input Quality
Classify PDF quality before analysis:
- 良好: Full text extractable
- 降级处理: Partial text, scanned sections, garbled encoding
- 严重降级: Minimal text, image-only PDF
If degraded: state reason in header line, proceed with available content, mark all gaps explicitly. Never fabricate content to fill gaps.
Optional: if user has Python, suggest running scripts/extract_pdf_meta.py
first for structured metadata.
Step 2: Classify Paper Type
Read references/paper-type-rubric.md and classify. Do NOT assume AI/ML.
Output the type label and 2–3 evidence indicators before proceeding.
Step 3: Execute Analysis
Follow references/output-schema.md for the selected mode. Apply all rules
from references/quality-checklist.md throughout every section.
Step 4: Self-Check Before Output
Verify before finalizing:
- Every uncertain field marked
[不确定]or[未明确给出] - Every contribution tagged
[原文声明]or[模型归纳] - No section silently omitted — skipped sections state why
- Paper type label matches rubric evidence
Anti-Hallucination Rules
Full rules in references/quality-checklist.md. Non-negotiable constraints:
- Source tagging:
[原文声明]= directly stated in paper (cite location);[模型归纳]= inferred by model (state reasoning basis) - Uncertainty:
[未明确给出]when absent;[不确定]when ambiguous - No domain assumption: classify paper type first, always
- No fabrication: venue, DOI, year, affiliations not in text →
[未明确给出] - Evidence binding: each contribution must cite section/figure/table/quote
- Degraded PDF: state which sections were unreadable; do not fill gaps
Degraded Input Fallback
| Situation | Action |
|---|---|
| Only abstract available | quick mode, note limitation |
| Scanned PDF, no text | Ask user for text or OCR first |
| Missing references section | Skip prior work analysis, note absence |
| Figures unreadable | Skip figure analysis, note absence |
| Non-English paper | Translate key sections, note source language |
Extended Mode: Author Prior Work
Only in extended mode:
- Extract all author names from paper
- Identify self-citations in reference list (shared authors)
- Infer research group focus from affiliations + paper title
- List prior works from reference list only — no web search, no external knowledge
- Tag all output:
[基于论文内引用，非外部检索] - If insufficient info: explicitly state "信息不足，无法判断前作关系"
Presentation Mode: PPT Generation
Only in presentation or presentation_with_figures mode.
Step A: Collect Overrides
Before building the slide plan, check if the user specified any of:
audience(lab / conference / general) — default:labduration_hint(10min / 20min / 30min) — default:20mintalk_style(technical / overview / discussion) — default:technicalemphasis(which sections to expand)skip(which sections to omit)
If not specified, use defaults silently.
Step B0: Extract PDF Figures (presentation_with_figures only)
Before building the slide plan, run:
python scripts/extract_pdf_figures.py <pdf_path>
This saves all figures to figures/ and writes figures/index.json with name, path, and page for each image. Use this index when assigning figure_ref paths in the handoff.
Step B: Build Slide Plan
Follow references/presentation-schema.md for structure.
Follow references/presentation-style-guide.md for compression rules.
- Map each slide role to the corresponding output-schema section
- Apply user overrides (emphasis → expand, skip → omit)
- For
presentation_with_figures: setfigure_needed: trueon method/result slides where a figure is the primary evidence; addfigure_refandfigure_hint - Slide count from duration_hint (10min→6-7, 20min→9-10, 30min→12-14)
Step C: Call pptx Skill
Follow references/pptx-handoff.md for the exact handoff format.
- Strip all
[原文声明]/[模型归纳]tags before passing to pptx - Do NOT include speaker notes in the handoff
- Call pptx skill automatically — do not ask the user first
- Exception: if user said "只要大纲" / "just the outline", output the slide plan as text and skip pptx
Source: flyer-Li/paper-analyst — distributed by TomeVault.