Agent Skills by Allen Pan
Showing 71 vetted skills indexed across 2 GitHub repositories.
HuskyDannyblogwatcher
Automate blog and RSS feed monitoring with blogwatcher-cli.
HuskyDannypolymarket
Retrieve prediction market data from Polymarket's public REST APIs.
HuskyDannyllm-wiki
Develop a Markdown-based knowledge base from human-curated sources.
HuskyDannyresearch-paper-writing
Automates ML/AI research-paper writing and management workflows from experiment design to submission preparation.
HuskyDannyxitter
Post, read, search, and manage tweets via the x-cli terminal tool.
HuskyDannyjupyter-live-kernel
Executes stateful Python code in a live Jupyter kernel environment.
HuskyDannysystematic-debugging
Guide developers through a four-phase root cause debugging process.
HuskyDannyrequesting-code-review
Automate pre-commit code review with static security scanning and auto-fix loops.
HuskyDannydspy
Program language models declaratively and optimize prompts with dspy.
HuskyDannygrpo-rl-training
Implement GRPO fine-tuning of language models with TRL and custom reward functions.
HuskyDannypeft-fine-tuning
Fine-tune large language models with LoRA and QLoRA for parameter-efficient training.
HuskyDannyaxolotl
Automate language model fine-tuning with Axolotl configurations and training.
HuskyDannypytorch-fsdp
Optimize PyTorch FSDP training with parameter sharding, mixed precision, and CPU offloading.
HuskyDannyunsloth
Fine-tune LLMs with LoRA and QLoRA using the Unsloth framework.
HuskyDannyaudiocraft-audio-generation
Generate music and sound effects from text using the PyTorch-based AudioCraft library.
HuskyDannysegment-anything-model
Segment objects in images using point, box, or mask prompts.
HuskyDannystable-diffusion-image-generation
Convert text descriptions into high-resolution images with Stable Diffusion.
HuskyDannymodal-serverless-gpu
Deploy ML models to Modal's serverless GPU cloud with Python-based configuration files.
HuskyDannygguf-quantization
Convert and quantize models to GGUF format for CPU and GPU inference.
HuskyDannyserving-llms-vllm
Serve LLMs with vLLM's PagedAttention and continuous batching.
HuskyDannyobliteratus
Remove refusal behaviors from open-weight LLMs using mechanistic interpretability techniques.
HuskyDannyoutlines
Generate structured text outputs matching JSON, XML, or code schemas.
HuskyDannyevaluating-llms-harness
Evaluate LLMs on academic benchmarks like MMLU and HumanEval.
HuskyDannyweights-and-biases
Automate MLOps workflows with Weights & Biases for experiment tracking and model versioning.