Skills
100,928 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Use when a user wants to set up, configure, install, or reconfigure the opencode Fusion agent team—a strong main/build agent that plans and reviews but cannot edit files, delegating all edits to a cheaper sidekick subagent, plus an explore search agent and optional research/design/reviewer/vision specialists. Triggers include “set up fusion,” “configure fusion,” “install fusion,” “fusion setup,” “undo fusion” / “remove fusion,” changing which models the main, sidekick, or explore agents use, or naming a subscription to start from a ready-made profile—for example, “set up fusion with my OpenCode Go subscription” (also OpenCode Zen, ChatGPT Plus/Pro, GitHub Copilot). Writes the global opencode config under ~/.config/opencode/.
Use when asked to “growth loops,” “build a growth engine,” “design a viral loop,” “create a content loop,” “move beyond paid acquisition,” or “why isn't growth compounding.” Helps design self-reinforcing growth systems where output becomes input. The Growth Loops framework (from Brian Balfour / Reforge and Elena Verna) shifts thinking from linear funnels to compounding loops.
Hunting skill for sqli vulnerabilities. Built from 12 public bug bounty reports including modern NoSQL injection (Rocket.Chat CVE-2021-22911 MongoDB $regex, Mongoose ORM CVE-2024-53900 $where bypass), modern ORM raw-fragment SQLi (Django CVE-2024-42005, Sequelize GHSA-wrh9-cjv3-2hpw), second-order SOQL injection (HackerOne Salesforce), time-based blind SQLi in GraphQL resolvers, and SQLi on OIDC-proxy backends. Use when hunting SQLi on any target. Dedicated NoSQL operator injection (MongoDB/CouchDB $where/$regex/$ne) is owned by hunt-nosqli — NoSQL appears here only as adjacent ORM/WAF context.
Teach concepts, principles, or evaluate viewpoints through experience rather than definitions: first present a situation suited to the learner's level and let them choose, then use a fair surprise to reveal a knowledge gap, name it only after structuring it, and cycle through transfer scenarios to retest. USE WHEN the user invokes '/ljg-teach', says 'teach me through scenarios', 'don't give me the answer first', 'help me learn through examples', or wants to understand a concept or viewpoint through choices and feedback. NOT FOR direct definition queries (use ljg-is/ljg-word), eight-dimensional concept analysis (use ljg-learn), or guided reading (use ljg-read).
Merge multiple fine-tuned models using mergekit to combine capabilities without retraining. Use when creating specialized models by blending domain-specific expertise (math + coding + chat), improving performance beyond single models, or experimenting rapidly with model variants. Covers SLERP, TIES-Merging, DARE, Task Arithmetic, linear merging, and production deployment strategies.
Deploy ML models on Kubernetes with KServe (formerly KFServing) and NVIDIA Triton Inference Server. Includes canary deployments, autoscaling, model versioning, A/B testing, and GPU resource management for production model serving
Use when building anything multi-step, anything that looks repeatable, anything the user mentions reusing, or any workflow with more than ~10 nodes. Triggers on “reuse”, “I do this in another workflow”, “extract”, “modular”, “shared logic”, “subworkflow”, multi-step builds, or any task that mentions logic the user has built before.
Implement OAuth 2.0 authentication with GitHub and Microsoft Entra (Azure AD) in Cloudflare Workers and other edge environments. Covers provider-specific quirks, required headers, scope requirements, and token handling without MSAL. Use when: implementing GitHub OAuth, Microsoft/Azure AD authentication, handling OAuth callbacks, or troubleshooting 403 errors in OAuth flows.
Retrieve detailed metadata and stored content for individual OpenRouter generations. Use when the user wants to inspect a specific request — its cost, latency, token usage, provider routing, or the actual prompt/completion text — or is debugging a failed or unexpected generation.
Detect potential code-level performance smells in Java — streams, collections, boxing, regex, object creation. Provides awareness, not absolutes — always measure before optimizing. For JPA/database performance, use jpa-patterns instead.
Use when asked to “position my product,” “positioning canvas,” “differentiate from competitors,” “figure out our category,” “repositioning,” or “why customers should pick us.” Helps define competitive alternatives, differentiated value, target customers, and market category. April Dunford's positioning framework from “Obviously Awesome” makes your product's value obvious to the right customers.
Build Remotion videos with snapcn — a shadcn registry of copy-paste animation components and timeline-driven UI primitives. Use when composing video scenes, adding text animations, transitions, backgrounds, UI blocks, brand/social cards, or full compositions in a Remotion project. Triggers include "snapcn", "video component", "add animation", "text reveal", "product demo video", "remotion component", "terminal simulator", "device frame", "phone mockup", "burned-in captions", "karaoke captions", "logo animation", "launch video", "changelog video", and the UI-primitive tier: "video input", "typing caret". Even if the user doesn't mention snapcn, activate when they need polished video primitives for Remotion.
Generates test files, creates mocking strategies, analyzes code coverage, designs test architectures, and produces test plans and defect reports across functional, performance, and security testing disciplines. Use when writing unit tests, integration tests, or E2E tests; creating test strategies or automation frameworks; analyzing coverage gaps; performance testing with k6 or Artillery; security testing with OWASP methods; debugging flaky tests; or working on QA, regression, test automation, quality gates, shift-left testing, or test maintenance.
Clinical variant interpretation, from raw variant calls to ACMG-classified recommendations with structural impact analysis. Use for VUS classification, pathogenicity assessment with cited criteria, structure-based variant impact (AlphaFold/PDB), non-coding/regulatory variant effect prediction with sequence deep-learning models (AlphaGenome, Enformer, Borzoi, ChromBPNet, Evo 2), and producing clinical-grade variant reports for return of results or molecular tumor boards. Use whenever a user asks about a variant's significance, an intronic/promoter/enhancer/UTR non-coding variant's functional impact, or needs ACMG classification — even if they don't say “ACMG”.
Use when you need to set up Java application profiling to detect and measure performance issues — including trusted preinstalled async-profiler v4.x setup, problem-driven profiling (CPU, memory, threading, GC, I/O), interactive profiling scripts, JFR integration with Java 25 (JEP 518, JEP 520), or collecting profiling data with flamegraphs and JFR recordings. This should trigger for requests such as Improve the code with profiling; Apply Profiling; Refactor the code with profiling; Add profiling support; Collect JFR or async-profiler data for Java performance. Part of Plinth Toolkit
Harden AI/LLM deployments against prompt injection, data exfiltration, model theft, and supply chain attacks. Covers input validation, output filtering, access control, model API security, and compliance controls for production AI systems.
Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components—not for simple single-file HTML/JSX artifacts.
Zero-config animations for React, Vue, Solid, Svelte, Preact with @formkit/auto-animate (3.28kb). Prevents 15 documented errors including React 19 StrictMode bugs, SSR imports, conditional parents, viewport issues, drag & drop conflicts, and CSS transform bugs. Use when: animating lists/accordions/toasts, troubleshooting SSR animation errors, React 19 StrictMode issues, or need accessible drop-in transitions with auto prefers-reduced-motion
Turns a task the owner keeps doing by hand into a named, reusable skill they can trigger by name or put on a schedule. Watches them do it once or listens to them describe it, works out what varies and what stays fixed, writes the skill with approval checkpoints in the right places, registers it with the router, and shares it with the team. This is how a business gets a workflow nobody shipped for them. Use this whenever the owner describes something repetitive they wish were automatic — including phrasings like "I do this every week," "can you remember how to do this," "make this a thing I can just ask for," "build me an agent," "automate this for me," "turn this into a workflow," or "I want this to happen every Monday." Reach for it when they describe a routine, even without asking for automation.
Store key-value data globally with Cloudflare KV's edge network. Use when: caching API responses, storing configuration, managing user preferences, handling TTL expiration, or troubleshooting KV_ERROR, 429 rate limits, eventual consistency, cacheTtl errors, wrangler types issues, or remote binding configuration
Write or edit a codependix.config.ts — configure which graph types run, per-project overrides, include and exclude globs, the whole-workspace graph, where each export lands, and the boundary rules applied to every built graph. Use when a workspace has no codependix configuration, a run produced no output because everything resolved to target none, adding a JSON or Markdown destination, choosing between an anchor block and a standalone file, a configuration is rejected for a missing destination or empty selector, adding a forbid, allow, or acyclic rule, registering the per-project Nx gate, or deciding whether one configuration can describe every project in a workspace.
Run a codependix dependency graph export or boundary check; choose between --check boundaries, --check reports, and --write; point a run at a workspace root or configuration file; or read a Mermaid block or JSON graph it produced. Use when running codependix or npx codependix, when a run exits 0 without writing anything, when --check is refused for having no value, when looking for a flag to select a graph type, when wiring codependix into a CI step or an Nx task, when a codependix-bound target fails, or when reading an exported Nx Neighborhood, NestJS module graph, or file-level import graph. Covers the command-line host directly, without assuming any task runner.
Answer dependency navigation and architecture questions using codependix CLI commands or committed graphs. Use when finding dependency paths between projects, files, or NestJS modules; tracing blast radius; determining what touches or imports a file or package; choosing narrowing flags to minimize run costs; or reading committed README diagrams across packages without running commands.
Conduktor platform expertise for Apache Kafka management, governance, and self-service. Covers Console (observe and manage), Gateway (enforce and proxy with interceptors), and CLI (operate and automate). Use when working with Conduktor configuration, deployment, Kafka data governance, encryption, multi-tenancy, or self-service workflows.
Create Databricks AI/BI dashboards. Must use when creating, updating, or deploying Lakeview dashboards, as Databricks Dashboard has a unique JSON structure. CRITICAL: You MUST test ALL SQL queries via CLI BEFORE deploying. Follow guidelines strictly.
Essential development workflow agents for code review, debugging, testing, documentation, and git operations. Includes 7 specialized agents with strong auto-discovery triggers. Use when: setting up development workflows, code reviews, debugging errors, writing tests, generating documentation, creating commits, or verifying builds
.NET and ASP.NET Core security patterns. Covers Identity, authentication, dependency auditing, secure coding practices, and OWASP for .NET ecosystem. USE WHEN: user works with "C#", ".NET", "ASP.NET Core", "Entity Framework", asks about ".NET vulnerabilities", "NuGet security", ".NET authentication", "Blazor security" DO NOT USE FOR: general OWASP concepts - use `owasp` or `owasp-top-10` instead, Java/Python security - use language-specific skills
Query Developer Experience (DX) data via the DX Data MCP server PostgreSQL database. Use this skill when analyzing developer productivity metrics, team performance, PR/code review metrics, deployment frequency, incident data, AI tool adoption, survey responses, DORA metrics, or any engineering analytics. Triggers on questions about DX scores, team comparisons, cycle times, code quality, developer sentiment, AI coding assistant adoption, sprint velocity, or engineering KPIs
Architect scalable backend systems, database schemas, APIs, and cloud infrastructure for robust server-side applications. Use when you need microservice vs monolith decisions, database indexing strategies, API versioning, event-driven architecture, ETL pipelines, WebSocket streaming, data modeling, query optimization, or cloud-native service design with high reliability and sub-20ms query performance.
Intelligently organizes your files and folders across your computer by understanding context, finding duplicates, suggesting better structures, and automating cleanup tasks. Reduces cognitive load and keeps your digital workspace tidy without manual effort.
Build a risk-matched whitebox or the approved production game by adapting to the host environment's available rendering stacks, 3D/2D generators, and modeling tools. Turn GAME_DESIGN and ART_DIRECTION into a concise BUILD_BRIEF and a runnable game candidate with deterministic self-checks. Use to prototype the riskiest design question, implement the approved game design, or turn the design into a running 2D, 2.5D, or 3D game. 游戏构建执行。结合本机与宿主环境实际可用的渲染栈与生成/建模工具（如 Three.js、混元 3D 等 3D 模型、Blender、2D/音视频工具），先针对头号设计风险搭建白盒原型，或将已批准的 GAME_DESIGN 与 ART_DIRECTION 实现为可运行游戏并配齐自检。用于验证核心设计风险、构建游戏原型或实现正式可玩版本。
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.