PDF & office files
3,236 rows, by stacks then stars
Run2 pdf form filling advanced
cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3-flash-preview/court-form-filling/run2_pdf-form-filling-advanced Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3-flash-preview/offer-letter-generator/run2_docx-expert Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3-flash-preview/schedule-planning/run2_pdf_parsing_pro Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run2_excel-weighted-mean Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run2_excel-stats Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run2_excel-verification Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b2-self-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run2_excel-formula-lookup Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 Extract PDF Form Field Names
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-haiku-4-5/court-form-filling/run3_Extract-PDF-Form-Field-Names Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 Save Modified Word Document to Disk
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-haiku-4-5/offer-letter-generator/run3_Save-Modified-Word-Document-to-Disk Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 Extract and Replace Text in Word Document Runs
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-haiku-4-5/offer-letter-generator/run3_Extract-and-Replace-Text-in-Word-Document-Runs Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 Extract PPTX Content and Structure
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-haiku-4-5/organize-messy-files/run3_Extract-PPTX-Content-and-Structure Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 Extract PDF Content and Metadata
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-haiku-4-5/organize-messy-files/run3_Extract-PDF-Content-and-Metadata Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 Extract DOCX Content and Structure
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-haiku-4-5/organize-messy-files/run3_Extract-DOCX-Content-and-Structure Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-haiku-4-5/schedule-planning/run3_pdf-calendar-parsing Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/court-form-filling/run1_python-pdf-form-filling Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run2 python docx nested table recursion
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/offer-letter-generator/run2_python-docx-nested-table-recursion Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/offer-letter-generator/run2_docx-template-filling Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 pdf text extraction for classification
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/organize-messy-files/run3_pdf-text-extraction-for-classification Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/schedule-planning/run1_pdf-calendar-parsing Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/weighted-gdp-calculation/run3_inspect-excel-structure Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 excel lookup formulas gnumeric compatible
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-opus-4-6/weighted-gdp-calculation/run3_excel-lookup-formulas-gnumeric-compatible Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-sonnet-4-6/court-form-filling/run3_inspect-pdf-fields Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-sonnet-4-6/offer-letter-generator/run3_fill-docx-template Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 organize files by subject
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-sonnet-4-6/organize-messy-files/run3_organize-files-by-subject Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-sonnet-4-6/schedule-planning/run1_pdf-calendar-parsing Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 excel lookup two conditions
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-sonnet-4-6/weighted-gdp-calculation/run1_excel-lookup-two-conditions Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 excel workbook cross sheet reference
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-sonnet-4-6/weighted-gdp-calculation/run1_excel-workbook-cross-sheet-reference Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 excel net exports percent gdp
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-claude-sonnet-4-6/weighted-gdp-calculation/run1_excel-net-exports-percent-gdp Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-flash-lite-preview/court-form-filling/run1_pdf-form-filling Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 docx template manipulation
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-flash-lite-preview/offer-letter-generator/run1_docx-template-manipulation Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-flash-lite-preview/schedule-planning/run3_pdf_calendar_parser Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 document content extraction
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-flash-lite-preview/organize-messy-files/run3_document_content_extraction Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-pro-preview/court-form-filling/run1_pypdf-form-filling Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 python docx advanced manipulation
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-pro-preview/offer-letter-generator/run1_python-docx-advanced-manipulation Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-pro-preview/weighted-gdp-calculation/run2_excel-2d-lookup-formulas Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run2 excel statistical and weighted mean formulas
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3.1-pro-preview/weighted-gdp-calculation/run2_excel-statistical-and-weighted-mean-formulas Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/court-form-filling/run3_inspect_pdf_form Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/court-form-filling/run3_fill_pdf_form Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/nlp-paper-reproduction/run2_extract_pdf_content Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run2 docx custom conditional logic
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/offer-letter-generator/run2_docx_custom_conditional_logic Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run2 docx recursive text replacement
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/offer-letter-generator/run2_docx_recursive_text_replacement Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 extract pdf text with poppler
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/organize-messy-files/run3_extract_pdf_text_with_poppler Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/organize-messy-files/run3_batch_file_organization Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run3 extract office text with pandoc
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/organize-messy-files/run3_extract_office_text_with_pandoc Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/schedule-planning/run3_analyze_pdf_calendar Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 excel weighted average sumproduct
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run1_excel-weighted-average-sumproduct Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run1_excel-two-way-lookup Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 excel descriptive statistics
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run1_excel-descriptive-statistics Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
Run1 excel percentage scaling rounding
cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/weighted-gdp-calculation/run1_excel-percentage-scaling-rounding Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-haiku-4-5/court-form-filling/pdf-form-filler Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-haiku-4-5/organize-messy-files/document-classifier Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-haiku-4-5/organize-messy-files/batch-file-organizer Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-haiku-4-5/schedule-planning/pdf-calendar-extraction Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-haiku-4-5/weighted-gdp-calculation/excel-financial-calculations Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-haiku-4-5/weighted-gdp-calculation/excel-two-condition-lookups Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-opus-4-6/court-form-filling/pdf-form-filler Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-opus-4-6/offer-letter-generator/docx-template-fill Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-opus-4-6/organize-messy-files/pdf-classifier Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-opus-4-6/organize-messy-files/document-organizer Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.
cxcscmu/SkillLearnBench/skills/b4-skill-creator-claude-opus-4-6/weighted-gdp-calculation/excel-weighted-stats Skill
75★ repo[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.