document-analysis

Reads and analyzes PDF, DOCX, XLSX, CSV, and other document formats with full-content extraction.

3|1|Updated Feb 14, 2026
One-click install
npx skills add https://github.com/LiboMa/agenticops-chat --skill document-analysis-liboma
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: document-analysis
Source: https://github.com/LiboMa/agenticops-chat/tree/main/skills/document-analysis
Command: npx skills add https://github.com/LiboMa/agenticops-chat --skill document-analysis-liboma

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pymupdf, pypdf, python-docx, openpyxl.

What problem does it solve? When users share documents like PDF reports, Word specs, or spreadsheets, standard file-reading tools truncate content at a few thousand characters, making deep analysis impossible. This Skill registers a read_document tool that returns complete document content, with page-range support for large PDFs, so nothing important is cut off. ## Core Features & Use Cases - Full-Content Reading: Reads PDF, DOCX, Markdown, HTML, CSV, JSON, YAML, and XLSX files without truncation, unlike the 4K-char-limited default file reader. - PDF Page Ranges: Reads specific pages of large PDFs (e.g., pages="1-5") to handle big reports in manageable chunks. - Guided Analysis Workflows: Built-in decision trees for explaining, summarizing, extracting from, and comparing documents, with tailored guidance for incident reports, architecture docs, cost reports, and spreadsheets. - Use Case: A user uploads a 40-page incident postmortem PDF and asks for the root cause and remediation steps. The agent reads the first pages for an overview, then targets the relevant sections, and returns a structured summary with timeline, root cause, and action items. ## Quick Start Read the attached report.pdf and summarize the key findings, conclusions, and recommended action items.

Frequently Asked Questions about document-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I read a large PDF file without truncation?▼

Use the read_document tool with the pages argument to read large PDFs in chunks, for example pages="1-5" for the first five pages. Start with an overview of the opening pages, then read specific sections as needed.

What file formats can be analyzed with read_document?▼

read_document supports PDF, DOCX, Markdown, HTML, CSV, JSON, YAML, and XLSX formats. PDFs use pymupdf or pypdf, Word files use python-docx, and Excel workbooks use openpyxl with multi-sheet support.

How is read_document different from read_local_file?▼

read_local_file truncates output at 4000 characters for operational safety, while read_document returns the complete document content. Use read_document when you need full understanding of a file rather than a quick preview.

Can I extract text from a scanned PDF document?▼

No, scanned PDFs contain only images with no text layer, so extraction returns empty content. The document needs OCR processing first, so ask for a text-based version or an OCR-processed copy.

Why does reading a password-protected PDF fail?▼

Password-protected PDFs cannot be opened by the underlying PDF libraries, so the read operation fails. Provide an unprotected copy of the file to analyze its contents.

What are the limits when analyzing large Excel spreadsheets?▼

Very large spreadsheets are truncated to 6000 characters of output even though all sheets and rows are parsed. Narrow the analysis by specifying columns of interest or filtering the data before reading.