cas-bigdatalab

199 mods across 1 repository, 541 stars between them.

en_text_cleaner

73

cas-bigdatalab/piflow

Skill Claude CodeCodex

An English text cleaner that expands common abbreviations, fixes common spelling errors, and can standardize capitalization in JSONL data.

not rated 541 2d ago A 104 tokens

encoding_converter

74

cas-bigdatalab/piflow

Skill Claude CodeCodex

A file-conversion tool that detects text encodings and converts files or directories to a chosen encoding, such as UTF-8.

not rated 541 2d ago A 110 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

A document-analysis tool that extracts and summarizes selected attributes of a person, subject, or other named entity across multiple documents.

not rated 541 2d ago A 135 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

A text filter that checks samples for specified sensitive or prohibited words and filters those exceeding a chosen threshold.

not rated 541 2d ago A 98 tokens

format_detector

77

cas-bigdatalab/piflow

Skill Claude CodeCodex

A file-format detection tool that identifies whether resources are text, spreadsheets, documents, images, HTML, data, or an unknown type. It also recommends a processing route for each resource.

not rated 541 2d ago A 167 tokens

frontend-design

78

cas-bigdatalab/piflow

Skill Claude CodeCodex

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build vue-web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any vue-web…

not rated 541 2d ago A 79 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

A text-normalisation tool that converts full-width letters, numbers, and punctuation to standard half-width characters. This is useful in Chinese, Japanese, and Korean text, where both forms may appear.

not rated 541 2d ago A 118 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

A file-conversion tool for Gaussian checkpoint files. It uses Gaussian’s formchk program to turn a binary .chk file into a readable formatted checkpoint file.

not rated 541 2d ago A 30 tokens

gaussian_executor

81

cas-bigdatalab/piflow

Skill Claude CodeCodex

A tool that runs Gaussian, a program for quantum-chemistry calculations, to perform structure optimisation, frequency analysis, and single-point energy calculations.

not rated 541 2d ago A 49 tokens

cas-bigdatalab/piflow

Skill Claude Code

A geographic conversion tool for turning image-segmentation results into map locations. Image segmentation means marking parts of an image, while geographic conversion links those marks to real-world coordinates.

not rated 541 2d ago A 47 tokens

gjf_editor

83

cas-bigdatalab/piflow

Skill Claude CodeCodex

A tool for creating and editing Gaussian input files, which describe quantum-chemistry calculations. It sets calculation options, resource limits, charge, and spin state.

not rated 541 2d ago A 69 tokens

gully_slop_operator

84

cas-bigdatalab/piflow

Skill Claude Code

A terrain-analysis tool that calculates slope features for gullies, which are channels cut into the ground by running water. It can use slope information from a digital elevation model, a grid representing land height.

not rated 541 2d ago A 42 tokens

cas-bigdatalab/piflow

Skill Claude Code

A tool for calculating sensitivity to flash floods and hydrological hazards. Sensitivity analysis estimates how prone an area is to a particular risk.

not rated 541 2d ago A 43 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image-quality filter that selects or rejects images according to their aesthetic score, meaning an assigned measure of visual appeal.

not rated 541 2d ago A 98 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image filter that keeps samples whose width-to-height ratio falls within a chosen range. The aspect ratio is the relationship between an image's width and height.

not rated 541 2d ago A 94 tokens

image_deduplicator

88

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image deduplication tool that finds exact duplicate images using image hashes and removes duplicate samples at the document level.

not rated 541 2d ago A 97 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image filter that keeps samples containing a chosen number of detected faces. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 101 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image filter that keeps samples where detected faces occupy a chosen share of the image area. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 106 tokens

image_nsfw_filter

91

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image-safety filter that keeps samples with a low NSFW score, where NSFW means content that may be unsuitable for work or general audiences. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 102 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image-pair filter that keeps pairs whose visual similarity falls within a chosen range. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 115 tokens

image_shape_filter

93

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image-shape filter that keeps images whose width and height fall within chosen limits. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 112 tokens

image_size_filter

94

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image-file-size filter that keeps dataset images within a chosen byte, KB, or MB range. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 112 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image-and-text matching filter that keeps items whose matching score falls within a chosen range. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 110 tokens

cas-bigdatalab/piflow

Skill Claude CodeCodex

An image-and-text relevance filter that keeps dataset items within a chosen similarity range. It requires a Python environment with the Data-Juicer package installed.

not rated 541 2d ago A 111 tokens

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: