Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/kujiangmudao/tablepack/pdf-table-to-excelnpx skills add kujiangmudao/tablepack --skill pdf-table-to-excelgit clone --depth 1 https://github.com/kujiangmudao/tablepackWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/kujiangmudao/tablepack/pdf-table-to-excel)<a href="https://agentmods.dev/skills/kujiangmudao/tablepack/pdf-table-to-excel"><img src="https://agentmods.dev/badge/skills/kujiangmudao/tablepack/pdf-table-to-excel.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00132 | $0.01415 |
| Opus 5 | $0.00066 | $0.00707 |
| Sonnet 5 | $0.00026 | $0.00283 |
| Haiku 4.5 | $0.00013 | $0.00142 |
Grade A, and why
pdf-table-to-excel scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TablePack — PDF tables → Excel packages
Read root AGENTS.md first. This skill is the executable checklist.
Product name TablePack; CLI module remains python -m pdf_excel.
Who is the user?
| Path | Action before convert |
|---|---|
| A — already has MinerU | Confirm mineru on PATH or config.yaml → mineru_bin. Do not require the user to learn MinerU CLI — only run TablePack. |
| B — no MinerU | Windows: scripts/install_mineru.ps1 · Unix: scripts/install_mineru.sh. Activate .venv-mineru. See docs/INSTALL.md. Still no MinerU tutorial — only python -m pdf_excel. |
Product promise: install MinerU (or Path B), put PDFs in pdf/, one command packages Excel + 原始表格/ + notes.
Model requirement (critical)
| Stage | Needs vision? | Why |
|---|---|---|
Run python -m pdf_excel |
No | CLI + MinerU HTML path |
| QC / fix sheets | Yes — multimodal | Open 原始表格/*.jpg and compare to Excel |
If the model cannot see images:
- Still run packaging.
- Do not claim “已严格质检”.
- Ask for a vision-capable model or human review for QC.
Environment (portable)
Prefer config.yaml / env / CLI — do not hardcode another machine’s paths.
| Item | Default |
|---|---|
| Entry | python -m pdf_excel or python convert_pipeline.py |
| PDFs | pdf/ |
| Output | output/ |
| MinerU cache | work/mineru_raw/ |
| MinerU | config.yaml / MINERU_BIN / PATH / .venv-mineru |
Recommended MinerU flags (also set in config):
mineru -p <pdf> -o work/mineru_raw -b pipeline -m auto -l ch -t true -f false
CPU-only → pipeline. Higher-accuracy VLM/hybrid only if installed and requested; QC still mandatory.
Required package layout
output/<PDF_STEM>/
├── <PDF_STEM>.xlsx
├── 原始表格/ # required screenshots
├── 图片/
└── 转换说明.md | 问题说明.md
Business rules
- MinerU is primary parser; OCR/pdfplumber only for repair.
- One PDF → one Excel (same stem).
- All tables → sheets in that one workbook.
- Always include
原始表格/images for human comparison. - After auto convert: visual QC vs screenshots; rewrite bad sheets.
- Unrecoverable → markdown explanation; never invent cells.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 159 lines · 132 tokens per session scan A 4df29b28004a
pdf-table-to-excel is a skill published in the GitHub repository kujiangmudao/tablepack (5 stars, last pushed 23d ago), licensed MIT. It adds 132 tokens to every session and 1,415 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
document-processing
Use when the deliverable is a document's bytes or its literal content — text/tables out of PDFs, AcroForm fill and flatten, page merge/split, PDF/DOCX from templates, OCR of image-only scans. NOT schema-typed fields pulled from text (that is structured-extraction), NOT signature routing (e-signature) or spreadsheet…
document-processing-expert
Read, generate and modify office documents and PDFs from code: PDF extraction and forms, Word documents, Excel workbooks and PowerPoint decks. Use when the user mentions PDF, DOCX, XLSX, PPTX, Word, Excel, PowerPoint or spreadsheets, wants data extracted from documents, needs a report or invoice generated as a file…
office-mcp
MCP server with 39 tools for Word, Excel, PowerPoint, PDF, OCR operations.
paddleocr-mcp
Use the local PaddleOCR MCP SDK v2 server for image OCR, PDF/document parsing, screenshots, tables, and structured text extraction.
deepread-pii
Redact PII from documents before sharing or sending to LLMs. 14 PII types (names, SSN, credit cards, medical records, etc.) detected with context-aware AI — not regex. Knows patient vs. doctor, personal vs. institutional. Black bar redaction on PDFs, scanned images, and text files. Free tier: 2,000 pages/month.
deepread-form-fill
AI-powered PDF form filling via DeepRead. Upload any PDF form + your data as JSON — AI detects fields, maps data semantically, fills the form with quality checks, returns a completed PDF. Works with scanned, non-editable forms — no AcroForm required.