Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.
82
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Evaluation results
92%
87%Employee Onboarding Form Completion
Fillable PDF form workflow
check_fillable_fields run
0%
100%
extract_form_field_info run
0%
100%
field_info.json structure correct
0%
100%
convert_pdf_to_images run
0%
0%
field_values.json created
0%
100%
field_values format correct
0%
100%
field_ids match extracted fields
0%
100%
All 8 employee fields covered
0%
100%
Values populated with real data
0%
100%
fill_fillable_fields run
33%
100%
Without context: $0.3867 · 1m 36s · 20 turns · 23 in / 6,333 out tokens
With context: $0.6303 · 1m 52s · 26 turns · 28 in / 7,883 out tokens
100%
100%Vendor Registration Form Completion
Non-fillable PDF form annotation workflow
check_fillable_fields run
0%
100%
convert_pdf_to_images run
0%
100%
fields.json created
0%
100%
fields.json pages dimensions
0%
100%
fields.json bounding box format
0%
100%
No intersecting bounding boxes
0%
100%
check_bounding_boxes run
0%
100%
create_validation_image run
0%
100%
fill_pdf_form_with_annotations run
0%
100%
Without context: $0.4695 · 1m 55s · 21 turns · 27 in / 7,142 out tokens
With context: $2.0321 · 8m 10s · 48 turns · 1,369 in / 32,280 out tokens
65%
Sales Data Extraction and Executive Summary Report
PDF table extraction and report generation
pdfplumber import
100%
100%
extract_tables() called
100%
100%
pandas DataFrame used
0%
0%
First row as column headers
0%
0%
to_excel() used
0%
0%
pypdf NOT used for extraction
100%
100%
reportlab used for summary PDF
100%
100%
Summary PDF created
100%
100%
Without context: $0.6852 · 2m 34s · 31 turns · 37 in / 10,756 out tokens
With context: $0.7543 · 2m 16s · 28 turns · 32 in / 8,179 out tokens
- Evaluated
- Agent
- Claude Code
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.