OpenParable
Skip to content

Development skills

Browse reusable Agent Skills, each with a clear purpose and practical guidance.

create-package

Create or modify gluestack-core and gluestack-utils packages for component creators and utilities

5.17k repo stars227 repo forks
Development

enhance-component

Improve or extend an existing gluestack-ui component with new features, variants, or fixes

5.17k repo stars227 repo forks
Development

gluestack-ui-v4

Enforces constrained, opinionated styling patterns for gluestack-ui v4. Main overview skill that coordinates specialized sub-skills for setup, components, styling, variants, performance, and validation.

5.17k repo stars227 repo forks
Development

gluestack-ui-v4:setup

Guide for installing gluestack-ui v4 per the official Installation doc - CLI and Manual paths only. Follow https://v4.gluestack.io/ui/docs/home/getting-started/installation strictly.

5.17k repo stars227 repo forks
Development

migrate-to-v5

Migrate a gluestack-ui project from v2, v3, or v4 to v5 (NativeWind v5 or UniWind). Covers automated CLI upgrade and manual step-by-step migration for all source versions.

5.17k repo stars227 repo forks
Development

vibe-coding

Transform an AI agent into a tasteful, disciplined development partner. Not just a code generator, but a collaborator with professional standards, transparent decision-making, and craftsmanship. Use for any development task: building features, fixing bugs, designing systems, refactoring. The human provides vision and decisions. The agent provides execution with taste and discipline.

5.17k repo stars766 repo forks
Development

devmoji

Devmoji commit message emoji conventions. Use when writing git commit messages, formatting commit descriptions, or choosing the right emoji prefix for a commit. Triggers on "commit message", "git commit", "commit emoji", "devmoji".

5.16k repo stars537 repo forks
Development

add_hf_model

Add support for a HuggingFace model in XTuner so it can be trained with the full set of parallelism strategies and optimization switches. The user can provide a HuggingFace Hub repo id, a local model directory, or a model family already supported by `transformers`. This skill walks through the full path: locating the reference implementation, classifying the model, splitting new code across the model / module / ops layers, implementing the XTuner model class and config, registering the entry point, validating bitwise numerical parity with HuggingFace, writing regression tests, and enabling the training optimizations (EP/SP, micro-batch, torch.compile, fp8, activation offload).

5.16k repo stars430 repo forks
Development

model_normalize

Normalize an HF model checkpoint — quantize bf16/fp16 weights to per-block FP8, and/or repack engine-named safetensors into HF-standard ~4GB shards.

5.16k repo stars430 repo forks
Development

model_patch

Merge a base HF model dir with an extra HF model dir by index diff — take tensors only present in extra, append them to a new output dir in HF-standard shard layout. Base wins on overlap.

5.16k repo stars430 repo forks
Development