This skill provides Microsoft Office file processing capabilities, using a progressive disclosure mechanism.
Supported formats
- Excel (.xlsx, .xlsm, .csv, .tsv) - Spreadsheet creation, editing, data analysis
- PowerPoint (.pptx) - Presentation creation, editing, design
- PDF (.pdf) - Document extraction, merging, form filling
- Word (.docx) - Document creation, editing, revision tracking
Progressive disclosure
This skill uses an on-demand loading mechanism, where detailed documentation is read only when needed, avoiding occupation of initial tokens.
Context efficiency
Traditional approach:
- All detailed content loaded into SKILL.md
- Estimated context: ~5000 tokens
This skill's approach:
- Metadata only: ~150 tokens
- Detailed documentation (on demand): ~5k tokens
On-demand loading guide
When you need to work with Office files in a specific format, read the corresponding detailed documentation:
# Excel 处理
cat $SKILL_DIR/references/xlsx.md
# PowerPoint 处理
cat $SKILL_DIR/references/pptx.md
# PDF 处理
cat $SKILL_DIR/references/pdf.md
# Word 处理
cat $SKILL_DIR/references/docx.md
Note: Replace $SKILL_DIR with the actual discovered path of this skill directory.
Usage pattern
When a user request matches this skill's capabilities:
Step 1: Identify the file type（Excel / PowerPoint / PDF / Word）
Step 2: Load the corresponding detailed documentation
cd $SKILL_DIR
cat references/<type>.md
Step 3: Perform operations according to the guidance in the detailed documentation
Skill source
This skill integrates Anthropic's official skills:
- xlsx - Excel processing
- pptx - PowerPoint processing
- pdf - PDF processing
- docx - Word processing
Original source:https://github.com/anthropics/skills
This skill uses progressive disclosure to minimize initial context usage