Wayner Barrios
Community@waybarrios · United States
CS-EEng
Agent Skills by Wayner Barrios
Showing 54 vetted skills indexed across 1 GitHub repositories.
waybarrioshuggingface-papers
Fetches Hugging Face paper pages and arXiv metadata via the papers API.
waybarrioshuggingface-local-models
Selects and runs GGUF models from Hugging Face locally with llama.cpp.
waybarriosfeature-dev
Guides feature implementation through a seven-phase workflow with codebase exploration and review.
waybarrioshuggingface-llm-trainer
Train and fine-tune language models with TRL or Unsloth on Hugging Face Jobs.
waybarriossecurity-threat-model
Generate repository-grounded AppSec threat models covering assets, trust boundaries, abuse paths, and mitigations.
waybarrioshuggingface-lora-space-builder
Build and publish Gradio Hugging Face Spaces for image and video LoRA demos.
waybarriostransformers-js
Run Hugging Face models in JavaScript with Transformers.js across browser, Node.js, Bun, and Deno.
waybarrioshuggingface-community-evals
Run local GPU evaluations of Hugging Face Hub models using inspect-ai and lighteval.
waybarrioshf-cloud-sagemaker-iam-preflight
Discover, validate, or create a SageMaker IAM execution role before deploying models.
waybarriosvuln-report
Drafts disclosure-ready GitHub security advisory reports for confirmed vulnerabilities.
waybarriostrl-training
Train and fine-tune transformer language models using TRL CLI commands.
waybarriossemgrep-rule-creator
Creates and tests custom Semgrep rules for detecting security vulnerabilities and code patterns.
waybarrioshuggingface-zerogpu
Implement and debug Hugging Face Spaces ZeroGPU code using the @spaces.GPU decorator.
waybarriossemgrep
Run Semgrep static analysis scans across codebases and merge findings into SARIF reports.
waybarrioshf-cloud-sagemaker-deployment-planner
Plans Amazon SageMaker model deployments by selecting inference pathways and instance types.
waybarriosdifferential-review
Performs security-focused differential review of code changes with git history analysis and markdown reports.
waybarrioshuggingface-spaces
Build, deploy, debug, and maintain Hugging Face Spaces using Gradio, Docker, or Static SDKs.
waybarriosinsecure-defaults
Detects fail-open insecure defaults like hardcoded secrets and weak authentication in application code.
waybarrioshuggingface-paper-publisher
Publish, index, and link arXiv research papers on Hugging Face Hub.
waybarrioscode-review
Reviews pull requests and code changes for bugs and convention violations using multi-agent detection.
waybarrioshf-cli
Operate the Hugging Face Hub CLI for authentication, repositories, jobs, endpoints, and spaces.
waybarrioshf-cloud-serving-image-selection
Selects and verifies region-specific SageMaker serving container URIs from the AWS Deep Learning Containers catalog.
waybarriossarif-parsing
Parses, filters, deduplicates, and aggregates SARIF static analysis results from security scanners.
waybarriostrain-sentence-transformers
Train and fine-tune SentenceTransformer bi-encoders, CrossEncoder rerankers, and SparseEncoder models.
Frequently Asked Questions About Wayner Barrios
FAQPage SchemaWhat tasks can I accomplish with Wayner Barrios's skill collection?▼
You can train and fine-tune LLMs with TRL or Unsloth, run GGUF models locally via llama.cpp, build Gradio Spaces, deploy models to SageMaker, analyze research papers, and run security audits using Semgrep, CodeQL, threat modeling, and vulnerability disclosure reporting.
Who is the target audience for these skills?▼
ML engineers fine-tuning and serving Hugging Face models, AppSec engineers performing static analysis and variant hunting, and software developers seeking structured feature design, code review, and architecture planning across Rust, TypeScript, and Python codebases.
How do I run models locally using these skills?▼
The huggingface-local-models skill guides GGUF selection, quantization choices, and llama.cpp server setup on CPU, Mac Metal, CUDA, or ROCm, including OpenAI-compatible serving. The hf-mem skill estimates memory required to load Safetensors or GGUF weights before downloading.
Are these skills free and open source?▼
Yes. All 54 skills are open source under Apache-2.0 or MIT licenses, with modifications documented in UPSTREAMS.json. There is no cost to use them, though underlying services like Hugging Face Jobs, SageMaker, and ZeroGPU may incur their own platform charges.
What prerequisites are needed for the SageMaker deployment skills?▼
You need an active AWS profile, region, and caller identity (verified via hf-cloud-aws-context-discovery), an isolated Python environment with boto3, and a valid SageMaker execution IAM role. The deployment planner skill coordinates image selection, endpoint configuration, and production defaults.