ML Ops
Machine learning operations
context-degradation
by shipshitdev
Recognize, diagnose, and mitigate patterns of context degradation in agent systems. Use when context grows large, agent performance degrades unexpectedly, or debugging agent failures.
advanced-evaluation
by shipshitdev
Master LLM-as-a-Judge evaluation techniques including direct scoring, pairwise comparison, rubric generation, and bias mitigation. Use when building evaluation systems, comparing model outputs, or establishing quality standards for AI-generated content.
claude-usage
by tdimino
Report ground-truth Claude token consumption and estimated cost by parsing JSONL session files directly. Use when checking API spend, auditing token usage by project/session/model, generating daily/weekly/monthly cost reports, or diagnosing ccusage undercounting. Includes subagent files that ccusage ignores.
parakeet
by tdimino
Local speech-to-text via Handy app (push-to-talk) and NeMo CLI scripts. Parakeet V3: 25 languages, auto-detection, ~30x realtime on M4 Max, 6% WER. This skill should be used when transcribing audio files or dictating voice input.
llama-cpp
by tdimino
Secondary local LLM inference engine via llama.cpp. This skill should be used when running GGUF models directly, loading LoRA adapters for Kothar, benchmarking inference speed, or serving models via llama-server. Complements Ollama (which remains primary for RLAMA and general use).
digital-twin-modeling
by kishorkukreja
When the user wants to build digital twins, simulate supply chain operations, or create virtual replicas for testing. Also use when the user mentions "digital twin," "simulation modeling," "discrete event simulation," "system simulation," "virtual supply chain," "what-if analysis," "scenario simulation," or "operational testing." For network optimization, see network-design. For forecasting, see demand-forecasting.
israeli-pension-advisor
by skills-il
Navigate the Israeli pension and savings system including pension funds (keren pensia), manager's insurance (bituach menahalim), training funds (keren hishtalmut), and retirement planning. Use when user asks about Israeli pension, "pensia", "keren hishtalmut", retirement savings, "bituach menahalim", pension contributions, or tax benefits from savings. Covers mandatory pension, voluntary savings, and withdrawal rules. Do NOT provide specific investment recommendations or fund performance comparisons.
woodpecker-cli
by fred-drake
Reference for the Woodpecker CI command-line tool. Use when working with Woodpecker CI pipelines, managing repositories, secrets, registries, organizations, or users via the CLI. Covers pipeline operations (start, stop, approve, logs), repository management, secret/registry configuration, and local pipeline execution.
ux-kano-model
by geekatron
"Kano model feature classification and prioritization sub-skill for the /user-experience parent skill. Classifies product features into Must-be (M), Performance (O), Attractive (A), Indifferent (I), and Reverse (R) categories using the functional/dysfunctional questionnaire pair methodology (Kano et al., 1984). Computes Customer Satisfaction (CS) coefficients (Better/Worse) for priority matrix visualization. Produces feature classification reports, priority matrices, and survey design templates. Sample size awareness: 5-8 respondents yields directional classification only (MEDIUM confidence); 20+ respondents required for statistical classification (Berger et al., 1993). Invoked by ux-orchestrator during Wave 4 lifecycle-stage routing or when user intent is \"Need to prioritize features\" at any lifecycle stage. Triggers: Kano, must-be, attractive, one-dimensional, performance feature, satisfaction, feature classification, delighter, feature prioritization, CS coefficient."
mineru-extract
by blessonism
Use the official MinerU (mineru.net) parsing API to convert a URL (HTML pages like WeChat articles, or direct PDF/Office/image links) into clean Markdown + structured outputs. Use when web_fetch/browser can’t access or extracts messy content, and you want higher-fidelity parsing (layout/table/formula/OCR).
codex-cto
by tdimino
Inverted orchestration—Codex CLI acts as CTO (planning and reviewing), Claude Code executes with native tools. Use when a task benefits from external architectural oversight before and after implementation.
smolvlm
by tdimino
Local vision-language model for image analysis using SmolVLM-2B
creating-financial-models
by ynulihao
This skill provides an advanced financial modeling suite with DCF analysis, sensitivity testing, Monte Carlo simulations, and scenario planning for investment decisions
Health & Nutrition Expert
by frankxai
Apply cutting-edge 2025 nutrition science on longevity, metabolic health, gut microbiome, and evidence-based dietary patterns for optimal vitality and disease prevention
Gym & Training Expert
by frankxai
Apply cutting-edge exercise science from 2025 research on hypertrophy, progressive overload, biomechanics, and evidence-based training protocols for optimal strength and muscle development
data-ml
by baz-scm
Competence in data analytics and machine learning, enabling developers to build data-driven features and integrate AI/ML capabilities.
3d-web-experience
by boisenoise
"Expert in building 3D experiences for the web - Three.js, React Three Fiber, Spline, WebGL, and interactive 3D scenes. Covers product configurators, 3D portfolios, immersive websites, and bringing ..."
computer-vision-warehouse
by kishorkukreja
When the user wants to apply computer vision in warehouses, detect defects, track packages, count inventory, or automate visual inspection. Also use when the user mentions "computer vision," "image recognition," "object detection," "barcode reading," "package tracking," "quality inspection," "YOLO," "RCNN," "image classification," or "warehouse automation with vision." For general ML, see ml-supply-chain.
mcp-memory-gateway
by IgorGanapolsky
The Agentic Feedback Studio & Veto Layer. Persistent agent memory, high-density context packs, and Agentic Guardrails (V2V) for Claude Code, Codex, and Gemini.
discover-ml
by rand
Automatically discover machine learning and AI skills when working with machine learning, PyTorch, training, inference, RAG, embeddings, fine-tuning, LLM, DSPy, HuggingFace, or diffusion models. Activates for ML development tasks.
Ollama Mastery Skill
by markus41
Ollama Model Library
voice-ai
by ScientiaCapital
"Production voice AI agents with sub-500ms latency. Groq LLM, Deepgram STT, Cartesia TTS, Twilio integration. No OpenAI. Use when: voice agent, phone bot, STT, TTS, Deepgram, Cartesia, Twilio, voice AI, speech to text, IVR, call center, voice latency."
unsloth-training
by ScientiaCapital
"Fine-tune LLMs with Unsloth using GRPO or SFT. Supports FP8, vision models, mobile deployment, Docker, packing, GGUF export. Use when: train with GRPO, fine-tune, reward functions, SFT training, FP8 training, vision fine-tuning, phone deployment, docker training, packing, export to GGUF."
sales-revenue
by ScientiaCapital
"B2B sales - cold outreach, lead scoring, pipeline metrics, MEDDIC/BANT qualification, discovery calls, and demos. Use when writing cold emails, scoring leads, running pipeline reviews, or preparing for sales calls."