trackio-cli — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited trackio-cli (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
A curated collection of Agent Skills for AI coding agents — covering technical blogging, web engineering, HuggingFace ML workflows, and full-stack AI research engineering. Compatible with Claude Code, GitHub Copilot, Cursor, Gemini CLI, and OpenAI Codex.
Skills follow the agentskills.io open standard.
This repository contains 114 production-ready skills organized into 5 collections:
| Collection | Skills | Domain |
|---|---|---|
devto_skills/ | 3 | Technical blog writing for dev.to |
engineering_skills/ | 5 | React, Next.js, mobile, UI/UX, Postgres |
huggingface_skills/ | 8 | HuggingFace Hub, model training, datasets |
airesearch_skills/ | 85 | Full AI research lifecycle (21 categories) |
wordpress_skills/ | 13 | WordPress plugin/theme/block development |
Each skill is a folder containing a SKILL.md file. When an AI agent starts in this workspace, it reads AGENTS.md at the root to discover all available skills and their trigger conditions. The agent loads individual skill instructions only when a relevant task is detected, keeping context usage minimal.
AGENTS.md <- AI reads this first (entry point)
devto_skills/
blog-writer/
SKILL.md <- Loaded when writing a dev.to article
rules/ <- Detailed guidelines (loaded on demand)
metadata.json
engineering_skills/
react-best-practices/
SKILL.md
rules/ <- 57 individual rule files
...Skills for writing high-quality technical blog posts on dev.to.
| Skill | Description |
|---|---|
blog-writer | SEO-optimized article writing with frontmatter, structure templates, title formulas, tag selection, and publish checklist |
thumbnail-generator | Prompts for AI image generators to create 1000x420px cover images in hand-drawn infographic style |
illustration-generator | Prompts for in-article diagrams — concept maps, workflows, architecture diagrams, comparisons |
Curated skills from Vercel Engineering and Supabase covering React, Next.js, mobile, UI/UX, and Postgres.
| Skill | Rules | Description |
|---|---|---|
react-best-practices | 57 | React/Next.js performance optimization across 8 categories: waterfalls, bundle size, server/client rendering, re-renders |
composition-patterns | — | React architecture patterns: compound components, state lifting, context/reducer, polymorphic APIs |
react-native-skills | 16 | React Native and Expo best practices: FlashList, Reanimated, safe areas, platform-specific patterns |
web-design-guidelines | 100+ | UI/UX audit: accessibility, focus states, forms, animation, typography, images, dark mode, i18n |
supabase-postgres-best-practices | 32 | Postgres optimization: query performance, connection pooling, RLS, schema design, indexing |
Skills for the complete HuggingFace ecosystem — from Hub CLI operations to model training and evaluation.
| Skill | Description |
|---|---|
hugging-face-cli | Download/upload models and datasets, manage repos, run cloud jobs via the hf CLI |
hugging-face-datasets | Create and manage datasets on HF Hub, define configs, SQL-based transformation |
hugging-face-model-trainer | Fine-tune LLMs using TRL (SFT, DPO, GRPO), GGUF conversion, hardware cost estimation |
hugging-face-evaluation | Add eval results to model cards, import from Artificial Analysis API, run vLLM/lighteval |
hugging-face-jobs | Execute Python scripts on HF infrastructure, manage scheduled jobs, monitor status |
hugging-face-trackio | Track ML experiments with Trackio, log metrics, build real-time dashboards on HF Spaces |
hugging-face-paper-publisher | Publish research papers on HF Hub, link to models/datasets, claim authorship |
hugging-face-tool-builder | Build reusable scripts for HF API operations, chain API calls, automate repeated tasks |
Production-ready skills covering the full AI research lifecycle, maintained by Orchestra Research.
| Category | Skills |
|---|---|
| Model Architecture | LitGPT, Mamba, NanoGPT, RWKV, TorchTitan |
| Tokenization | HuggingFace Tokenizers, SentencePiece |
| Fine-Tuning | Axolotl, LLaMA-Factory, PEFT, Unsloth |
| Mechanistic Interpretability | TransformerLens, SAELens, pyvene, nnsight |
| Data Processing | NeMo Curator, Ray Data |
| Post-Training | TRL, GRPO, OpenRLHF, SimPO, verl, slime, miles, torchforge |
| Safety & Alignment | Constitutional AI, LlamaGuard, NeMo Guardrails, Prompt Guard |
| Distributed Training | DeepSpeed, FSDP, Accelerate, Megatron-Core, Lightning, Ray Train |
| Infrastructure | Modal, Lambda Labs, SkyPilot |
| Optimization | Flash Attention, bitsandbytes, GPTQ, AWQ, HQQ, GGUF |
| Evaluation | lm-eval-harness, BigCode Eval, NeMo Evaluator |
| Inference Serving | vLLM, TensorRT-LLM, llama.cpp, SGLang |
| MLOps | Weights & Biases, MLflow, TensorBoard |
| Agents | LangChain, LlamaIndex, CrewAI, AutoGPT |
| RAG | Chroma, FAISS, Pinecone, Qdrant, Sentence Transformers |
| Prompt Engineering | DSPy, Instructor, Guidance, Outlines |
| Observability | LangSmith, Phoenix |
| Multimodal | CLIP, Whisper, LLaVA, BLIP-2, SAM, Stable Diffusion, AudioCraft |
| Emerging Techniques | MoE, Model Merging, Long Context, Speculative Decoding, Distillation, Pruning |
| ML Paper Writing | LaTeX templates, citation verification |
| Research Ideation | Research Brainstorming, Creative Thinking |
Skills from the WordPress core team covering plugin/theme development, blocks, REST API, WP-CLI, performance, and more. Targets WordPress 6.9+.
| Skill | Description |
|---|---|
wordpress-router | Classifies any WP repo and routes to the correct workflow/skill automatically |
wp-block-development | Gutenberg block development: block.json, attributes, dynamic rendering, deprecations |
wp-block-themes | Block themes: theme.json, templates, patterns, style variations, Site Editor |
wp-plugin-development | Plugin architecture, hooks, Settings API, security, cron, release packaging |
wp-rest-api | REST endpoints, WP_REST_Controller, schema validation, permission callbacks |
wp-interactivity-api | Interactive blocks with data-wp-* directives and @wordpress/interactivity store |
wp-performance | Backend profiling: WP-CLI doctor/profile, query optimization, caching, cron |
wp-wpcli-and-ops | WP-CLI: db, search-replace, multisite, cron, scripting with wp-cli.yml |
wp-playground | Disposable WP instances in browser or locally via @wp-playground/cli, blueprints |
wp-phpstan | PHPStan static analysis: phpstan.neon, baselines, WP-specific typing |
wp-abilities-api | WP Abilities API: wp_register_ability, REST exposure, permission checks |
wp-project-triage | Inspect any WP repo and generate a structured JSON triage report |
wpds | Build UIs using WordPress Design System (WPDS) components and tokens |
Skills work automatically with the following AI agents when this repository is opened as a workspace or cloned locally:
| Agent | How it loads skills |
|---|---|
| Claude Code | Reads AGENTS.md automatically on startup |
| GitHub Copilot (VS Code) | Reads AGENTS.md automatically |
| Cursor | Reads AGENTS.md as project rules |
| Gemini CLI | Reads AGENTS.md automatically |
| OpenAI Codex | Reads AGENTS.md automatically |
Clone this repository and open it as your workspace:
git clone https://github.com/nghidanh2005/HieuNghi-AI-Skills.git
cd HieuNghi-AI-SkillsOpen in VS Code:
code .AI agents will automatically discover skills via AGENTS.md when the workspace is active.
See CONTRIBUTING.md for the full guide on adding skills, including:
rules/ subfolders for large skill setsAGENTS.md in 3 places (skill table, decision tree, structure tree)HieuNghi-AI-Skills/
├── AGENTS.md <- Central AI entry point
├── README.md <- This file
├── CONTRIBUTING.md <- Guide for adding new skills
├── LICENSE <- MIT
├── devto_skills/
│ ├── blog-writer/
│ │ ├── SKILL.md
│ │ ├── metadata.json
│ │ └── rules/
│ ├── thumbnail-generator/
│ │ ├── SKILL.md
│ │ └── metadata.json
│ └── illustration-generator/
│ ├── SKILL.md
│ └── metadata.json
├── engineering_skills/
│ ├── react-best-practices/
│ ├── composition-patterns/
│ ├── react-native-skills/
│ ├── web-design-guidelines/
│ └── supabase-postgres-best-practices/
├── huggingface_skills/
│ ├── hugging-face-cli/
│ ├── hugging-face-datasets/
│ ├── hugging-face-model-trainer/
│ ├── hugging-face-evaluation/
│ ├── hugging-face-jobs/
│ ├── hugging-face-trackio/
│ ├── hugging-face-paper-publisher/
│ └── hugging-face-tool-builder/
└── airesearch_skills/
├── 01-model-architecture/
├── 02-tokenization/
├── 03-fine-tuning/
├── 04-mechanistic-interpretability/
├── 05-data-processing/
├── 06-post-training/
├── 07-safety-alignment/
├── 08-distributed-training/
├── 09-infrastructure/
├── 10-optimization/
├── 11-evaluation/
├── 12-inference-serving/
├── 13-mlops/
├── 14-agents/
├── 15-rag/
├── 16-prompt-engineering/
├── 17-observability/
├── 18-multimodal/
├── 19-emerging-techniques/
├── 20-ml-paper-writing/
└── 21-research-ideation/Skills in this repository are curated from:
MIT — See LICENSE for details.
Individual skill collections may reference libraries with separate licenses. See each upstream repository for details.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.