SkillsBench v1.1

SkillsBench Suite Explorer

Browse SkillsBench tasks through the benchmark's own taxonomy, then compare official scores across models, runtimes, and `with-skill` versus `no-skill` execution modes.

Visible Tasks 4
Official Runs 0
Ranked Profiles 0
Skill Mode All Modes

Task Explorer

Browse the public roster by taxonomy

Medium skillsbench/edit-pdf

Edit PDF

Category
Office White Collar
Subcategory
PDF Editing
Task Type
Transformation · Formatting
Modality
PDF · Document
Interface
Terminal · Python
Skill Type
File Format Knowledge · Tool Workflow

Task-scoped Skills PDF Editing · Document Updates

Outputs 1 Open Match
Medium skillsbench/data-to-d3

Data to D3

Category
Software Engineering
Subcategory
Data Visualization Frontend
Task Type
Implementation · Generation
Modality
CSV · Source Code · Webpage
Interface
Terminal · Browser
Skill Type
Library Api Usage · Tool Workflow

Task-scoped Skills D3 Charting · Frontend Data Viz

Outputs 4 Open Match
Medium skillsbench/weighted-gdp-calc

Weighted GDP Calc

Category
Finance Economics
Subcategory
Spreadsheet Analysis
Task Type
Calculation · Analysis
Modality
Spreadsheet
Interface
Spreadsheet App · Python
Skill Type
Mathematical Method · Data Cleaning Procedure

Task-scoped Skills Spreadsheet Formulas · Macro Finance

Outputs 1 Open Match
Easy skillsbench/court-form-filling

Court Form Filling

Category
Office White Collar
Subcategory
Legal Form Filling
Task Type
Extraction · Generation
Modality
PDF
Interface
Terminal · Python
Skill Type
Domain Procedure · File Format Knowledge

Task-scoped Skills Legal Form Filling · PDF Forms

Outputs 1 Open Match

Official Summary

Top combinations under the current slice

Official verification is still warming up, so there are no aggregate rows yet.

Recent Runs

Latest SkillsBench reports uploaded to the arena

No SkillsBench runs have been uploaded yet.

Difficulty Slice

Official score and pass rate by difficulty

No difficulty slice data yet.

Category Slice

Official score and pass rate by category

No category slice data yet.