cxcscmu/SkillLearnBench

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

83Stars on the repository
200Mods indexed here, across every type
1mo agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Common operations and patterns for using the xlsx skill tool to manipulate Excel files programmatically.

not rated 83 +1 1mo ago A 23 tokens original MIT

anthropic-brand

74

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Anthropic's official brand color palette, typography, and design tokens for use in branded materials.

not rated 83 +1 1mo ago A 23 tokens original MIT

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Creating technical diagrams and exploded-view illustrations using Python Pillow with geometric primitives and text annotations.

not rated 83 +1 1mo ago A 24 tokens original MIT

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Modern Mandarin rhyme groups for classical Chinese poetry composition, mapping finals to rhyme categories.

not rated 83 +1 1mo ago A 23 tokens original MIT

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Define custom distance/similarity metrics for DBSCAN clustering with sklearn, using weighted Euclidean distances.

not rated 83 +1 1mo ago A 25 tokens original MIT

parallel-processing

79

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Parallel processing with joblib for grid search and batch computations across multiple CPU cores.

not rated 83 +1 1mo ago A 19 tokens original MIT

pareto-optimization

80

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Multi-objective optimization with Pareto frontiers for finding trade-off solutions between conflicting objectives.

not rated 83 +1 1mo ago A 23 tokens original MIT

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Generate structured CSV security audit reports from vulnerability data with proper filtering and formatting.

not rated 83 +1 1mo ago A 22 tokens original MIT

geojson-processing

83

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Loading and processing GeoJSON earthquake and plate boundary data with GeoPandas for spatial analysis.

not rated 83 +1 1mo ago A 22 tokens original MIT

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Analyzing earthquake data relative to tectonic plate boundaries using PB2002 dataset and GeoPandas.

not rated 83 +1 1mo ago A 26 tokens original MIT

json-data-analysis

85

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Patterns for analyzing and outputting structured JSON answers from enterprise data queries.

not rated 83 +1 1mo ago A 18 tokens original MIT

13f-data-analysis

86

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Analyze SEC 13-F filings data including AUM, holdings count, and cross-quarter comparisons using TSV files.

not rated 83 +1 1mo ago A 27 tokens original MIT

fuzzy-search

87

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Fuzzy string matching for fund names and stock CUSIPs using thefuzz library.

not rated 83 +1 1mo ago A 22 tokens original MIT

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Apache Druid JavaScript RCE vulnerability (CVE-2021-25646) fix via @JacksonInject hardening and constructor validation.

not rated 83 +1 1mo ago A 36 tokens original MIT

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Preventing Jackson @JacksonInject override attacks where JSON input can replace server-side injected values.

not rated 83 +1 1mo ago A 24 tokens original MIT

gh-cli-pr-stats

90

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Computing pull request statistics (merge time, top contributors) from gh CLI JSON output using jq.

not rated 83 +1 1mo ago A 26 tokens original MIT

gh-cli-search

91

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Using the gh CLI to search and list GitHub issues and pull requests with date filters and label queries.

not rated 83 +1 1mo ago A 25 tokens original MIT

jq-processing

92

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Using jq to parse, filter, and aggregate JSON data from CLI tools for generating reports.

not rated 83 +1 1mo ago A 21 tokens original MIT

simpo-loss

94

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

SimPO (Simple Preference Optimization) loss computation for LLM alignment without a reference model.

not rated 83 +1 1mo ago A 22 tokens original MIT

docx-conditionals

95

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Handle conditional sections in Word templates with IF/ENDIF markers, removing or keeping content based on data values.

not rated 83 +1 1mo ago A 27 tokens original MIT

file-classification

96

cxcscmu/SkillLearnBench

Skill Claude CodeCodex

Classify academic papers and documents into subject categories using keyword-based text analysis.

not rated 83 +1 1mo ago A 19 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: