Skills for your AI
Optimizes email campaigns for mobile devices — responsive layout, images, typography, CTAs, links, subject lines, testing and analytics. Use when designing, revising, or testing mobile email templates, or analyzing mobile campaign performance.
Analyzes mobile game performance, assets, memory, battery, network, UI/UX, analytics, A/B tests, and monetization to produce prioritized optimization recommendations. Use when a developer asks to find bottlenecks, cut file size, reduce battery or data use, imp
Designs and refines mobile learning apps with AI-assisted content, features, and UX guidance, covering interface, content structure, gamification, interactivity, multimedia, progress tracking, social learning, accessibility, performance, and specialized learni
Produces structured mobile QA plans, checklists, and test cases across test creation, compatibility, usability, performance, security, regression, automation, accessibility, localization, and compliance. Use when a tester asks for test cases, a test plan, a ch
Analyzes and improves mobile website performance across technical, content, UX, and local SEO, producing prioritized recommendations and implementation guides. Use when the user asks for a mobile site audit, mobile keyword or voice search research, mobile read
Creates, updates, and audits Obsidian Maps of Content so vault notes stay navigable and nothing is orphaned. Use when folders lack a MOC, after large imports or edits, when images have no incoming links, or when auditing the MOC hierarchy.
Runs Python code in serverless Modal containers with GPUs, autoscaling, images, schedules, volumes, and web endpoints. Use when deploying ML models, running batch or parallel jobs, attaching GPUs, building dependency images, scheduling tasks, persisting data,
Implements, fine-tunes, pretrains, quantizes, and deploys LLMs with LitGPT's pretrained architectures and LoRA/QLoRA workflows. Use when the user wants to load a LitGPT model for inference, fine-tune on their own dataset, pretrain from scratch, prepare tokeniz
Guides installation, loading, configuration, benchmarking, and troubleshooting of Mamba-1 and Mamba-2 state-space models for sequence modeling and generation. Use when the user asks to install mamba-ssm, load a pretrained Mamba model, configure Mamba blocks, b
Guides users through training and sampling from nanoGPT, covering Shakespeare character-level training, GPT-2 124M reproduction, fine-tuning from pretrained checkpoints, custom datasets, troubleshooting, and architecture explanations. Use when the user asks ab
Explains the RWKV architecture and guides installation, inference, Transformer comparisons, and fine-tuning. Use when the user asks how RWKV works, wants to install or set up RWKV, run or stream inference, compare RWKV with Transformers, or fine-tune an RWKV m
Configures and launches PyTorch-native torchtitan pretraining jobs with FSDP2, TP, PP, CP, Float8 and SLURM. Use when the user wants to pretrain a model like Llama 3.1 or DeepSeek V3, set up single-node or multi-node training, enable Float8 or torch.compile, r
Evaluates AI models end-to-end with classification and regression metrics, cross-validation, bias audits, robustness probes, interpretability, resource and baseline comparisons, dashboards, and transfer or active learning plans. Use when a user supplies predic
Benchmarks AI models against success criteria, budget, latency and compliance constraints to select the best one. Use when choosing between candidate models, designing an evaluation, or confirming a quality regression in a deployed model.
Builds a modern minimalist one-page A4 resume from the user's work history, including print-ready styling. Use when the user wants to create, format, or export a resume, CV, or print/PDF-ready one-page document.
Analyzes a project's codebase exhaustively, documents every feature, and produces a modernization plan with architectural recommendations. Use when the user asks to modernize a project, document an existing codebase, identify its tech stack and architecture, o
Designs complete interactive training module blueprints covering content, scripts, storyboards, activities, assessments, feedback, simulations, and delivery formats. Use when a training goal needs a module plan, script, storyboard, quiz, role-play, simulation,
Converts molecular SMILES strings into numerical feature vectors using molfeat featurizers (fingerprints, 2D descriptors, pretrained embeddings, or combined). Use when the user provides SMILES and asks for ECFP/MACCS/MAP4/FCFP fingerprints, RDKit or Mordred de
Gathers full context from Monday.com for a bug item and produces a production-quality fix and pull request. Use when given a Monday.com bug item ID such as MON-1234, or when asked to fix, analyze, or open a PR for a Monday.com bug.
Automate Monday.com workflows, board management, team collaboration, and cross-board integrations
Brainstorm 3-5 monetization strategies with audience fit, risks, and validation experiments. Use when exploring revenue models, evaluating pricing strategies, or deciding how to monetize a product.
Analyzes MongoDB performance through database stats, logs, query and aggregation review, index and schema analysis, and explain benchmarking to produce actionable optimization recommendations. Use when asked to analyze MongoDB performance, review queries or ag
Monitors social platforms, news, and competitor activity to produce trend reports, sentiment analysis, influencer lists, and crisis alerts for media relations teams. Use when asked to track mentions or hashtags, summarize press coverage, compare competitor med
Guides systems administrators through planning and configuring IT monitoring across servers, networks, applications, logs, alerts, dashboards, and incident response. Use when setting up monitoring agents, network or APM tools, log pipelines, alert rules, dashb
