Skills
99,064 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
CodeStable is an AI coding workflow for serious software engineering that keeps humans in the loop. Rather than centering on Agent orchestration, it organizes requirements, architecture, features, issues, and historical decisions to make development driven by Codex and Claude controllable, traceable, and continuously evolvable.
Python error handling patterns for FastAPI, Pydantic, and asyncio. Follows "Let it crash" philosophy - raise exceptions, catch at boundaries. Covers HTTPException, global exception handlers, validation errors, background task failures. Use when: (1) Designing API error responses, (2) Handling RequestValidationError, (3) Managing async exceptions, (4) Preventing stack trace leakage, (5) Designing custom exception hierarchies.
Building or editing forms with @falcondev-oss/form — the type-safe, reactive, schema-driven form library (packages form-core, form-react, form-vue) built on @vue/reactivity. Use when working with useForm / useFormCore, form.fields accessors, $use(), field.value / field.handleChange / field.model, form.data, or a Standard Schema (zod v4 / arktype) form in a React or Vue codebase.
This skill should be used when picking or diagnosing a training move (SFT, LoRA, DPO/KTO/ORPO, RFT, GRPO/PPO/RLOO, RLHF), or when the user mentions fine-tuning, post-training, training recipe, reward design, or weight updates. Decision tree by reward shape, smoke-run gate, three failure diagnostics, five false-progress patterns. Provider recipes and I/O contract in references/.
Use when authoring or modifying a Datex Studio footprint-workflow (configurationTypeId=23, CLI type `footprintworkflow`, conventional -footprintWorkflow.json suffix) on a branch — a low-code TypeScript implementation that plugs into a named Footprint platform workflow extension point (Cartonization, Entity Status Change (Before Commit), Allocation Strategy, Recommend License Plate Location, Barcode Parser, …). Owns the platform-fixed signature contract (single `Input: FootPrintWorkflow.<Slot>InputBaseWL`, slot-dictated out-params), the slot binding (workflowDefinitionId/Name discovered via the workflowsMetadata API; workflowGUID is the code callers pass: generate fresh for a new workflow, preserve on edits, reuse the legacy value for a drop-in replacement), the action-tier calling rules, and the thin-dispatcher pattern. Triggers: "create/edit a footprint workflow", "implement the Cartonization workflow", "customize entity status change before commit", "allocation strategy workflow", "recommend location workflow", "what dxs commands manipulate Footprint workflows", "replace the legacy XAML Datex Workflow with TypeScript".
Create a Frappe Framework v15 DocType with proper controller, service layer, repository, and test files. Triggers: "create doctype", "new doctype", "frappe doctype", "add doctype", "/frappe-doctype". Generates production-ready DocType with type annotations, lifecycle hooks, and multi-layer architecture integration.
Read the approved PRD.md and generate the native sources for a third-party PAM control (the compiled `.ppmplugin` track) — iOS Obj-C `<Pascal>Module` plus optional system-frameworks podspec, Android Kotlin `<Pascal>Module` with build.gradle, AndroidManifest and ReactPackage, a dev-only private package.json (react + react-native devDeps for the builds), and the committed `./manifest.json` dispatch contract the PCF and build stage both read. No TypeScript INativeExtension layer — the contract is the manifest plus the native modules' dispatch surface. Emits the layout in shared/repo-layout.md and generates substantially complete native code (compiled later by /build-android-binary and /build-ios-binary, not here). Local only — writes files, runs no git and touches no remote or feed. PCF is generated by /generate-pcf-companion; the bundle is built by /generate-ppmplugin.
Generate the dispatcher PCF for a third-party `.ppmplugin` (wrap-runtime) control. Runs `pac pcf init` in the pcf/ subfolder, rewrites ControlManifest.Input.xml from ARCHITECTURE §6, and writes index.ts derived from ARCHITECTURE §4 (message contract), §8 (PCF surface), §9 (error UX) — no placeholders. The bridge dispatches the composite key `<name>/<receiver>` to `NativeModules.<nativeModule>.<method>` via the host-injected `window.PowerApps.NativeExtension.sendAsync` global (never `cordova.exec` — not in the PCF sandbox); also emits a `PowerAppsNativeExtension.d.ts` ambient declaration. Responses are peeled with `extractResponse`. Emits structured JSON debug/error logs. Validated by `npm run build`. **Local only** — does not deploy. Needs only `pac` CLI. Run after the native module exists. Uses npm (not pnpm).
Build and verify a third-party `.ppmplugin` end to end. This recommended entry point detects Android and iOS source, confirms target platforms, then invokes the manifest, platform build, assemble, and audit stages in order. Advanced users can invoke a stage directly for focused reruns or debugging. Produces a native-only bundle (`manifest.json` plus an Android DEX and/or iOS framework) and stops on any failed stage rather than emitting an unverified partial build. Run after the native module exists.
Validate, reconcile to the chosen target(s), and stage the `manifest.json` for a `.ppmplugin` bundle. In the normal flow the committed `./manifest.json` already exists (authored by /generate-native-extension), so this stage reads it, runs the plugin's upload-compatibility checks locally as a pre-flight gate (name regex, canonical prefix, known incompatible names, method and identifier shapes), reconciles `entrypoints` down to the platform(s) you ship, and writes the gitignored staged copy `ppmplugin/staging/manifest.json` that /assemble-ppmplugin zips. If no committed manifest exists (a hand-authored module) it falls back to deriving every field from the class name and Android module source, writing both the committed file and the staged copy. No toolchain — pure read, validate, write. Target-aware. Run after the native module exists, before the build skills.
USE FOR: GitHub: gh CLI usage, PR conflict resolution, commit messages, issue/PR creation, inline comments, sub-issues, review style, public surface path hygiene rules. Use this skill when tasks need this repository-specific workflow. DO NOT USE FOR: unrelated tasks, broad rewrites outside the request, or generated runtime outputs.
GitLab protected branch operations via API. ALWAYS use this skill when user wants to: (1) view branch protection rules, (2) protect/unprotect branches, (3) configure push/merge access levels, (4) set up code owner approval requirements.
Reduce the size of compiled Go binaries and container images: linker flags, inlining budget, CGO and build tags, embedded assets, dependency weight, and measuring what actually costs bytes. Use when a binary or image is too large, when shrinking a CLI for distribution, or when auditing what a dependency adds to the build. Trigger examples: "binary is too big", "shrink the binary", "reduce image size", "strip symbols", "what is making my binary large", "ldflags -s -w". Not for: runtime speed and allocations (go-performance-review), CI setup (go-ci), release pipelines and signing (go-build-release), dependency CVEs (go-dependency-audit)
Data persistence patterns in Go covering raw SQL with sqlx/pgx, ORMs like Ent and GORM, connection pooling, migrations with golang-migrate, and transaction management. Use when implementing database access, designing repositories, or managing schema migrations.
Audit Go nolint directives for staleness and lazy justifications. Mechanically verifies each suppression with golangci-lint, then runs adversarial Red/Blue/White debates on the top candidates for removal. Use when inheriting a Go codebase, during periodic cleanup, or when nolint count is growing unchecked.
Standardizes all click-related IDs and CSS classes across website for clean analytics tracking. Use when users want to "standardize analytics classes", "clean up tracking IDs", "prepare DOM for GTM", "fix analytics naming", or "make tracking consistent". Scans entire codebase (HTML/JSX/TSX/Vue) and applies consistent naming convention - IDs as "cta_{location}_{action}" and classes as "js-track js-{category} js-{action} js-{location}". Acts as senior frontend engineer ensuring scalable GA4/GTM implementation.
P1 phase skill for Headless Ghidra baseline and runtime evidence. Use after P0 when the target must be imported into Ghidra, auto-analyzed, exported to baseline YAML, and given reproducible runtime or hotpath observations without decompiling function bodies.
Implements search and filter interfaces for both frontend (React/TypeScript) and backend (Python) with debouncing, query management, and database integration. Use when adding search functionality, building filter UIs, implementing faceted search, or optimizing search performance.
When the user wants to submit an iOS app to the App Store. Use when the user mentions 'App Store,' 'App Store Connect,' 'TestFlight,' 'iOS submission,' 'app review,' 'EAS build,' 'eas submit,' 'Apple review,' 'app rejection,' or 'iOS release.' This skill covers the entire process from setup to App Store review approval.
Read Jira issue data from local cache or API, and download attachments via script (never into context). Triggers when user says "jira-read [JIRA-ID]", 「读取 Jira」「查看 Jira」「下载 Jira」「下载附件」 (read/view/download Jira / download attachments), or needs to fetch Jira issue data. Requires $JIRA_CACHE_DIR (e.g. ~/.cache/jira).
Expert in migrating monolithic, legacy systems to modern architectures using patterns like Strangler Fig, CDC, and Anti-Corruption Layers. Use when modernizing legacy codebases, breaking monoliths, or planning incremental migrations. Triggers include "legacy migration", "modernization", "strangler fig", "monolith to microservices", "legacy rewrite", "incremental migration".
ONLY when the user explicitly wants a thorough, exhaustive, or comprehensive multi-source investigation or report that can run for minutes — due diligence, market landscapes, "compare X across many entities", verified fact-finding. ALSO read this when writing code that integrates the /v1/research endpoint (modes, reasoning depths, async polling). Slower and more expensive than linkup-search. For normal lookups and research use linkup-search instead; for bulk records from a page use linkup-extract. Uses Linkup's async /v1/research REST endpoint. Requires LINKUP_API_KEY.
Guidance for .NET MAUI XAML data bindings, compiled bindings, value converters, binding modes, multi-binding, relative bindings, and MVVM best practices. USE FOR: "data binding", "compiled binding", "value converter", "IValueConverter", "binding mode", "TwoWay binding", "multi-binding", "relative binding", "BindingContext", "MVVM binding", "INotifyPropertyChanged". DO NOT USE FOR: CollectionView item templates (use maui-collectionview), Shell navigation data passing (use maui-shell-navigation), or dependency injection (use maui-dependency-injection).
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
Design and implement Minimal APIs in ASP.NET Core using handler-first endpoints, route groups, filters, and lightweight composition suited to modern .NET services. USE FOR: building new HTTP APIs in ASP.NET Core; creating lightweight microservices; choosing between Minimal APIs and controllers. DO NOT USE FOR: unrelated stacks; generic tasks that do not need this specific guidance. INVOKES: inspect the repository context, edit targeted files, and run relevant build, test, lint, or validation commands when changes are made.
Operation-aware node configuration guidance. Use when configuring nodes, understanding property dependencies, determining required fields, choosing between get_node detail levels, or learning common configuration patterns by node type. Always use this skill when setting up node parameters — it explains which fields are required for each operation, how displayOptions control field visibility, and when to use patchNodeField for surgical edits vs full node updates.
Proven workflow architectural patterns from real n8n workflows. Use when building new workflows, designing workflow structure, choosing workflow patterns, planning workflow architecture, or asking about webhook processing, HTTP API integration, database operations, AI agent workflows, batch processing, or scheduled tasks. Always consult this skill when the user asks to create, build, or design an n8n workflow, automate a process, or connect services — even if they don't explicitly mention 'patterns'. Covers webhook, API, database, AI, batch processing, and scheduled automation architectures. Also use when optimizing a slow workflow or speeding up large-item-count processing (node count, batchSize, all-items vs per-item).
Highest-priority for all NestJS-related tasks involving API engineering. Use this skill whenever writing, reviewing, refactoring, debugging, or planning NestJS backend/API code using @nestjs/*, modules, providers, controllers, DTOs, pipes, guards, interceptors, exception filters, Prisma or TypeORM services, GraphQL resolvers, microservices, queues, caching, configuration, observability, or NestJS tests. Prefer this skill over generic Node.js or backend advice for NestJS code.
Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations (for example: Codex, Responses API, Chat Completions, Apps SDK, Agents SDK, Realtime, model capabilities or limits); prioritize OpenAI docs MCP tools and restrict any fallback browsing to official OpenAI domains.
Open Orbit briefing skill — selected by the Orbit pipeline when Linear is the user's only connected connector, or when the user explicitly scopes their daily digest to Linear. Pulls the past 24 hours of issue movement, status changes, assignments, and cycle progress from the user's authenticated Linear connection and renders the digest in Linear's native Inbox + cycle-progress visual language. This skill should not be triggered manually — it is invoked by Orbit's daily-digest scheduler against live Linear data.
Use when generating SDKs for multiple languages from a single OpenAPI spec, or multiple SDK variants from different sources. Covers workflow.yaml multi-target configuration, per-language gen.yaml, monorepo structure. Triggers on "multiple SDKs", "multi-language SDK", "SDK for each language", "multi-target SDK", "SDK monorepo", "generate SDKs for".
Pachca — interactive forms and modal windows for bots. Creating forms, modal windows, polls, questionnaires and requests with input fields, dropdowns, checkboxes, date/time selection, file uploads. Use when the user wants to show a form, open a modal window, create a poll or questionnaire, handle form submission (view_submission), or validate fields. Also for trigger_id and dropdown/select. Requires a bot token. NOT for inline buttons in messages or configuring webhooks. Use when: показать форму, открыть форму, собрать форму, модальное окно, модалка, отправка формы, обработать форму, валидация формы, валидация поля, view_submission, опрос, анкета, заявка, форма заявки, form, modal, interactive form. NOT for: кнопки в сообщении, настроить бота, вебхук, → pachca-messages, → pachca-bots.
Create or revise PlantUML source for precise UML and architecture documentation, including class, sequence, component, deployment, state, activity, use-case, ER, ArchiMate, and C4 diagrams. Use when the user explicitly requests PlantUML, UML, C4, `.puml`, standard UML semantics, or detailed architecture notation. Prefer Mermaid for lightweight Markdown-native diagrams and ProcessOn only for hosted editable rendering.
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.