Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Discover all possible analysis domains for an artifact. Finds obvious and non-obvious angles — architecture, security, but also marketing positioning, user psychology, regulatory implications, teaching value. Use before prism-scan or prism-full to explore what's worth investigating.
Process raw meeting transcripts from Granola or other sources into structured notes with frontmatter, action items, summary, and formatted transcript. Use this skill when the user asks to process a meeting transcript or provides a raw transcript that needs formatting.
Analyze product or business data to support a decision or recommendation. Use when a decision depends on metric-backed evidence, such as choosing a direction, prioritizing an opportunity, evaluating a change, segmenting users, sizing tradeoffs, or deciding what to do next.
Guidance for designing baseline Microsoft Purview Data Loss Prevention (DLP) policies across Exchange, SharePoint, OneDrive, Teams, and Endpoint. Covers locations, conditions (SITs/labels/classifiers), severity-tiered actions, simulation mode, exception handling, Endpoint DLP onboarding, and tuning. WHEN: Purview DLP, data loss prevention policy, prevent data exfiltration, Endpoint DLP, DLP rule conditions, DLP simulation mode, block sensitive sharing, policy tips, DLP across Microsoft 365, tune DLP false positives.
Solves single- and multi-objective optimization problems in Python with pymoo, using NSGA-II, NSGA-III, MOEA/D, SPEA2, RVEA, GA, DE and PSO. Covers custom problems (Problem, ElementwiseProblem, FunctionalProblem), constraint handling, mixed-variable problems, ZDT/DTLZ/WFG benchmarks, genetic operators, parallel evaluation, Pareto front visualization and multi-criteria decision making. Use when finding Pareto-optimal trade-offs between conflicting objectives. Use when defining a constrained or mixed-variable problem and choosing an evolutionary algorithm. Use when benchmarking algorithms on standard test problems. Use when customizing crossover or mutation operators. Use when picking one solution from a Pareto front. Not for gradient-based or convex solvers.
Organizes PyTorch training code with the lightning package (PyTorch Lightning): LightningModule, LightningDataModule, Trainer, callbacks such as ModelCheckpoint and EarlyStopping, loggers (TensorBoard, W&B, MLflow, Comet, CSV), and multi-GPU/TPU strategies (DDP, FSDP, DeepSpeed). Use when structuring a PyTorch model into training, validation and test steps. Use when configuring a Trainer for multi-GPU or TPU runs. Use when writing a LightningDataModule for data loading. Use when adding checkpointing, early stopping or experiment logging to a training run. Use when choosing between DDP, FSDP and DeepSpeed for a model size. Not for plain PyTorch loops without Lightning.
Transform passive "Verify X" test descriptions into active, observable test actions. Use when test ideas lack specificity, use vague language, or fail quality validation. Converts to action-verb format for clearer, more testable descriptions.
Use when reporting progress in autonomous loop iterations. Triggers at the end of every autonomous loop iteration, when the autonomous-loop skill completes a BUILD phase, when progress reporting is needed for monitoring or exit evaluation, or when producing machine-parseable RALPH_STATUS blocks with exit signal protocol.
Create, inspect, or adapt an Obsidian knowledge base structure for content production (sources, knowledge, topics, writing, publishing, feedback loops). Use for “set up an Obsidian knowledge base”, “connect the vault to the content pipeline”, “inspect the knowledge base structure”, or when ray-writer cannot find the knowledge base. Only add missing structure; do not modify existing notes.
Turn ideas, clippings, research, or drafts into Chinese long-form articles with Ray’s perspective (WeChat Official Account / X long-form posts), integrated with the knowledge base’s finalization and publishing workflow; also create translated articles with authorized bylines. Use for “write as an article,” “write a WeChat Official Account or X long-form post,” “rewrite this,” or “translate and repost.” For spoken scripts, switch to ray-kb; does not publish.
Systematic code refactoring following Martin Fowler's catalog. Methodologies: characterization tests, Red-Green-Refactor, incremental transformation. Capabilities: SOLID compliance, DRY cleanup, code smell detection, complexity reduction, legacy modernization, design patterns, functional programming patterns. Actions: refactor, extract, inline, rename, move, simplify code. Keywords: refactor, SOLID, DRY, code smell, complexity, extract method, inline, rename, move, clean code, technical debt, legacy code, design pattern, characterization test, Red-Green-Refactor, functional programming, higher-order function, immutability, pure function, composition, currying, side effects. Use when: improving code quality, reducing technical debt, applying SOLID principles, fixing DRY violations, removing code smells, modernizing legacy code, applying design patterns.
Mid-conversation reflection skill that pauses execution and zooms out from detail mode to honestly reassess direction, assumptions, and bias. Use when the user says “reflect,” “take a step back,” “step back,” “zoom out,” “are we missing something,” “bigger picture,” “sanity check this,” “are we on track,” “are we overthinking this,” “forest for the trees,” or any variation signaling intent to break out of detail mode and reassess. Also trigger when the conversation has gone deep on implementation details without a strategic check-in, or when the user shows signs of being stuck — that’s often a signal the framing needs a reset, not more detail work. Intentionally low-intake: runs the 5-dimension analysis immediately when prior context is rich enough; asks one forcing clarifier only when invocation context is too thin to reassess from
Distills a logical relational-database design methodology into rules an agent applies while designing or reviewing a schema. Covers the design sequence (mission → tables → fields → keys → relationships → business rules → views → integrity review), one-subject-per-table decomposition, atomic single-valued fields, candidate/primary/foreign keys, relationship types with deletion rules and participation, the four levels of data integrity, database-vs-application business rules, validation tables, views for derived data, and the flat-file / spreadsheet / RDBMS-driven antipatterns to avoid. The structure is logical and RDBMS-agnostic, and normalized by construction. Use when designing a new relational schema, reviewing or refactoring an existing one, resolving redundant or repeating data, choosing keys, modeling relationships, or deciding where a constraint belongs.
Create finished vertical short videos with Remotion and post them to TikTok, Instagram Reels, and YouTube Shorts. Turns a topic, article, URL, dataset, changelog, or list into a 1080x1920 kinetic-text short (large animated words, stat count-ups, top-N lists, launch and changelog videos, quotes), timed to reading speed and kept inside each app's safe zone, rendered locally with no paid API. Use when someone asks for a Remotion short, a Remotion video for TikTok/Reels/Shorts, kinetic typography or text animation video, a stats or data short, a top 5 list video, a product launch or changelog video, a quote video, faceless text shorts, or “make a short about X and post it.” Includes a review sheet with app overlays marked, a cover frame, and WoopSocial posting with confirmation.
Review recent repository changes and decide whether AGENTS.md or other project-level documentation needs a high-level update. Use when finishing a feature, fix, refactor, or architectural change and you need to preserve repo-shaping guidance such as new patterns, constraints, workflows, validation rules, or onboarding-relevant gotchas without adding low-level implementation detail.
Review code changes with a small team — two senior engineers review independently, debate their findings, then a team lead reports a consolidated list. Local mode compares the current branch against a base branch; PR mode (auto-enabled when a PR URL or number is supplied) offers to post the findings as PR review comments. PR mode is strictly read-only — no file in the working tree is ever modified, and every developer instruction is interpreted as an update to the comments plan. The agent never checks out a branch in the current working directory; if it needs files at another ref, it creates a temporary worktree and removes it before presenting the report. Local mode lets the developer pick findings to fix, and the same two engineers implement them in auto mode. Engineers read the project's AI documentation (architecture first) before reviewing, and explicitly check for reuse violations and unused established practices.
Review the latest changes and check whether they comply with the project's documented guidelines (AGENTS.md, CLAUDE.md, or equivalent). Use when reviewing local diffs, recent commits, or feature work and you need a findings-first assessment of architecture, reuse, testing, and repo-specific rules.
Use for foundational Roblox experience development: deciding what runs on the client or server, where scripts and modules belong, how to structure reusable code, and how to handle everyday services, attributes, bindables, workspace objects, input, camera, raycasts, collisions, and CFrame-based gameplay scripting in Studio.
Use for Roblox multiplayer communication across the client-server boundary: designing RemoteEvent, UnreliableRemoteEvent, and RemoteFunction flows; validating client requests; handling replication-aware gameplay; applying rate limits and anti-exploit checks; reasoning about network ownership, server-authority patterns, Input Action System use in authoritative gameplay, and streaming-sensitive multiplayer correctness
RouterOS scripting language and CLI configuration idioms for .rsc files and interactive commands. Use when: writing or reviewing RouterOS scripts, scheduler/netwatch/on-event snippets, idempotent CLI config, :local/:global/:foreach/:do syntax, [find] selectors, print as-value handling, script permissions, or when the user mentions .rsc, RouterOS script, scripting tips, or CLI config automation.
Real-world data cohort analysis pipeline for claims and EHR data — cohort identification with ICD-10, NDC, and CPT codes, medication adherence metrics (PDC/MPR), propensity score estimation and balance diagnostics, Kaplan-Meier survival curves, and Cox proportional hazards models. Use when the user mentions claims data, cohort identification, ICD-10 codes, NDC codes, CPT codes, medication adherence, PDC, MPR, propensity score, SMD balance, Kaplan-Meier, Cox model, Schoenfeld residuals, survival analysis on claims, or real-world evidence pipeline.
Detect Cross-Site Scripting (XSS) vulnerabilities in a codebase using a three-phase approach: recon (find HTML/JS/DOM sink sites), batched verify (trace user input to sinks in parallel subagents, 3 sink sites each), and merge (consolidate batch results). Requires sast/architecture.md (run sast-analysis first). Outputs findings to sast/xss-results.md. Use when asked to find XSS or cross-site scripting bugs.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.