Skills
101,174 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Generate professional shot lists from screenplays and scripts. Use when user uploads a screenplay (.fountain, .fdx, .txt, .pdf, .docx) or describes scenes for production planning. Parses scripts to extract scenes, helps determine camera setups, shot types, framing, and movement through collaborative discussion, then generates beautifully formatted PDF shot lists for production. Triggers include requests to create shot lists, plan shots, break down scripts for filming, or organize camera coverage.
Transforms documents containing theoretical knowledge or frameworks (PDFs, markdown, book notes, research papers, methodology guides) into actionable, reusable Claude Code skills using systematic reading methodology. Use when user mentions "create a skill from this document", "turn this into a skill", "extract a skill from this file", or when analyzing documents with methodologies, frameworks, or processes that could be made actionable.
Converts Opus-quality skills into deterministic Haiku-executable workflows via trace-driven distillation and cross-model validation. Triggers on: "distill this skill", "make this skill work on Haiku", "cross-model optimization", "optimize skill for cost". NOT for code simplification, use code-refiner.
Instructions for AI agents to use the local smart-search CLI for current web search, source-backed fact checking, URL fetching, site mapping, official/API/documentation search, deep research, and reproducible evidence instead of MCP tools or native web search.
SpacetimeDB development best practices for TypeScript server modules and client SDK. This skill should be used when writing, reviewing, or refactoring SpacetimeDB code to ensure optimal patterns for real-time, multiplayer applications. Triggers on tasks involving SpacetimeDB modules, tables, reducers, subscriptions, or React integration.
Turn fuzzy intent into shared understanding and acceptance criteria — specification as a conversation, run one question at a time, before any story is split or planned. Use when an idea, feature request, or problem statement has no agreed rules or examples yet ("let's spec this out", "what should this actually do?", "we need acceptance criteria"). Produces an example map and acceptance criteria written back into the team's own story artifact, plus candidate glossary terms and parked questions. The agent-facilitated round is a draft for review by the accountable owner and, when risk or shared ownership warrants it, the relevant human perspectives. For decision trees with no artifact, use grill-me where installed or this skill's one-question protocol; for tightening an existing artifact, see find-gaps; for slicing agreed work, see story-splitting.
ScriptRunner Connect (SR Connect, SRC), Adaptavist's code-first iPaaS: whether and how a service can be integrated, workspaces, connectors, managed APIs, event listeners, scheduled triggers, record storage, runtime limits, plans and hosting, and doing the work through the @sr-connect/cli. Use for any SRC, ScriptRunner Connect or Atlassian-integration question, and load references/ before running the CLI.
Consolidate duplicate files or directories into one canonical home after approval, held as an identity by this machine's filesystem: hardlink, symlink, junction, or bind mount, with bytes, ACLs, owner and mode, named streams, xattrs and kept timestamps preserved. Use when consumers need the same bytes at several paths and textual references cannot serve them, when asked to hardlink, symlink, junction or bind-mount duplicate copies, or to dedupe identical files or trees on disk without breaking their readers.
Communicate accessibility decisions, requirements, and value to stakeholders who aren't accessibility specialists. Use when presenting accessibility work to leadership, product managers, engineers, or anyone who needs to understand why accessibility decisions matter. Triggers on: stakeholder, business case, justify accessibility, explain accessibility, leadership, executive, ROI, why accessibility, cost of accessibility, persuade, convince, accessibility presentation.
Complete Stripe refund and dispute lifecycle handling. PROACTIVELY activate for: (1) charge.refunded handler design, (2) charge.dispute.created / charge.dispute.closed handlers, (3) Refund delta computation from event.data.previous_attributes.amount_refunded, (4) Dispute-hold past_due status management, (5) shouldRestoreStatus predicate with satisfies Record<Stripe.Dispute.Status, boolean>, (6) Credit-pack vs subscription refund differentiation, (7) Checkout Session lookup for refund proportion math, (8) Allowlist default-deny for external enums, (9) stripe.checkout.sessions.list({payment_intent}) pattern, (10) Dispute outcome branch logic (won / lost / warning_closed / prevented). Provides: full handler patterns for all three events, predicate examples, credit-pack vs subscription math, exhaustive switch patterns.
Use when executing multi-task plans where each task can be implemented independently by a subagent. Triggers when a plan has 3+ independent tasks, when speed of execution is important, when tasks have clear acceptance criteria suitable for delegation, or when two-stage review gates (spec compliance and code quality) are needed for iterative fix cycles.
DEPRECATED — archived and unmaintained; superseded by conclave (https://github.com/anfreire/conclave). Spawn and conduct Claude Code worker sessions in detached tmux panes — run Claude Code in the background, from any agent on any harness. Use when asked to orchestrate or delegate work to Claude Code sessions, spawn workers in tmux, fan tasks out across parallel sessions, run an implement-then-review loop, or supervise long unattended Claude Code runs. Workers are supervised through file signals — never by parsing the terminal.
Expert guidance on SwiftNIO best practices, patterns, and implementation. Use when developers mention: (1) SwiftNIO, NIO, ByteBuffer, Channel, ChannelPipeline, ChannelHandler, EventLoop, NIOAsyncChannel, or NIOFileSystem, (2) EventLoopFuture, ServerBootstrap, or DatagramBootstrap, (3) TCP/UDP server or client implementation, (4) ByteToMessageDecoder or wire protocol codecs, (5) binary protocol parsing or serialization, (6) blocking the event loop issues.
Provides empirical test protocols and metrics to validate whether hypothesized symmetries actually hold in data or models before committing to equivariant architecture. Includes invariance/equivariance testing, group structure verification, and distribution analysis under transforms. Use when testing invariance, validating equivariance, checking symmetry assumptions, debugging symmetry-related model failures, or needing data-driven validation before architecture decisions.
Skill that analyzes existing TAKT workflows and facets and proposes improvements. Performs structural validation of workflow YAML, consistency checks between facets, style guide compliance verification, detection of unused facets, and rule design optimization proposals. If execution logs (.takt/logs/*.jsonl) exist, it also performs log-based diagnostic analysis, reporting rule evaluation efficiency, loop hotspots, ABORT rate, and more. Analysis is based on the style guides and engine specifications under references/takt. Triggers: 「ワークフローを分析」「taktの設定を確認」「ファセットの品質チェック」「ワークフローのレビュー」「takt analyze」「ワークフローの改善提案」「ワークフローの整合性チェック」「taktの問題を見つけて」「ログを分析」「実行ログの診断」「taktのログを見て」「ルール評価の統計」「ai_fallbackの頻度」「ループの検出」
Skill for creating and editing individual TAKT facets (Persona/Policy/Instruction/Knowledge/Output Contract). Generates standalone files that comply with each facet's style guide. Uses the style guides and built-in facet set under references/takt as reference material to determine facet types, select templates, and perform quality checks. Triggers: 「ペルソナを作りたい」「ポリシーを追加」「インストラクションを書く」「ナレッジを定義」「出力契約を作成」「ファセットを編集」「takt facet」「レビュアーのペルソナ」「コーディングポリシー」
Skill for optimizing existing TAKT workflows (workflow YAML and facet set). Implements token consumption reduction, step consolidation, rule simplification, promotion of facet reuse, improved loop control, and parallelization proposals, and directly generates the optimized files. Diagnostic results from takt-analyze (static analysis, log diagnosis) can be used as input. This skill handles only "executing optimization"; diagnosis and analysis are delegated to takt-analyze. Based on the engine specifications and style guides under references/takt. Triggers: 「ワークフローを最適化」「taktの高速化」「ワークフローを軽くしたい」「トークンを減らしたい」「ステップを減らしたい」「takt optimize」「ワークフローの効率化」「ファセットを整理したい」「ワークフローをスリムにして」「taktのコスト削減」「ワークフローをシンプルにしたい」
Use when hierarchical task breakdown is needed, when dependency mapping between tasks is required, when effort estimation and parallelization planning is needed, or when creating work breakdown structures. Triggers on /decompose command, when complex tasks need to be broken into manageable subtasks, when critical path analysis is needed for scheduling, or when identifying tasks that can run concurrently.
Use when reviewing who can access which resources in a Teleport cluster, or whether that access is actually used — access list / ACL recertification, "who can reach this database/server", "what can this user access", attesting access for audit/compliance, and finding dormant or unused standing privileges. Covers `tctl access-review`, the `access_path` SQL query language, output semantics, and combining it with `tctl investigate`.
Best practices for the Terra API Streaming (Real-Time) API – live, roughly per-second biometric data from wearables over websockets. Use when building realtime streaming with Terra API, TerraRT SDKs, or wss://ws.tryterra.co; streaming live heart rate, steps, distance, acceleration, ECG, HRV, RR intervals, calories, MET, power, cadence, speed, location, or gyroscope; pairing a wearable over BLE or ANT+; producing sensor data from a mobile app; consuming a live stream on a backend; or streaming from an Apple Watch or Wear OS watch. Covers the producer/broker/consumer architecture, the websocket handshake and opcodes, token minting, close codes, replay/backfill, and per-platform SDK setup (iOS, Android, React Native, Flutter, Wear OS).
Integrate Three.js 3D scenes into Editframe compositions via addFrameTask. Scenes are pure functions of time, fully scrubable, and renderable to MP4. Use when creating 3D animations, WebGL content in compositions, or integrating Three.js with Editframe's timeline system.
Zero-shot time series forecasting with Google's TimesFM foundation model. Use for any univariate time series (sales, sensors, energy, vitals, weather) without training a custom model. Supports CSV/DataFrame/array inputs with point forecasts and prediction intervals. Includes a preflight system checker script to verify RAM/GPU before first use.
PyTorch Geometric (PyG) for graph neural networks — node/link/graph classification, message passing (GCN, GAT, GraphSAGE, GIN), heterogeneous graphs, neighbor sampling, and custom datasets. Use when working with torch_geometric, not for general NetworkX analytics or non-graph PyTorch models.
Analyse accessibility tradeoffs when a design decision improves accessibility for one group but may affect another, or when accessibility competes with other requirements. Use when facing conflicting accessibility needs, business pressure to reduce accessibility scope, or technical constraints that limit accessibility options. Triggers on: tradeoff, trade-off, compromise, accessibility vs performance, accessibility vs deadline, conflicting needs, which users, who do we prioritise, can't do both, accessibility scope.
Market signals → daily digest delivered before market open. Scans watchlist tickers for regime changes, technical setups, options flow anomalies, and IBKR position health — then delivers a prioritized action list. No IBKR distraction during 50-call BDR days. Use when: 'market digest', 'trading alerts', 'pre-market scan', 'watchlist check', 'what's setting up', 'trading morning brief', 'market open prep'.
Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.
Read your TikTok niche and audience from real data. Scan a hashtag for top videos with plays, likes, and comments, see which hooks and sounds are working, pull a profile's videos (yours or a competitor's), and read the commenters on a video (TikTok hides likers, so commenters are the signal). Powered by Apify, no login. Triggers on "what is trending on TikTok", "analyze this hashtag", "competitor videos", "read my comments", "which sounds are working". Not for writing a script (use tt-hook-scripter).
Apply, review, and explain the opinionated conventions in the TypeScript Style Guide. Use automatically only when TypeScript conventions are part of the requested task, including type modelling, discriminated unions, function APIs, variables and state, naming, source organization, React props, component APIs, state management, data flow, or tests. Do not use for styling or layout alone.
Use when an extension has to work with a newer or the current TYPO3 LTS, when a version bump breaks compatibility or leaves deprecated APIs behind, when upgrading v11->v12, v12->v13 or v13->v14 (v14.3 LTS is the current target), when one codebase must stay compatible with two versions, when running Extension Scanner, Rector, Fractor or PHPStan against a target version, or when a specific v14 breaker bites - Fluid 5 strict ViewHelpers, HashService removal, the ext_tables.php split.
Hardens TYPO3 v14 installations and extensions with secure configuration, trusted hosts, file permissions, Install Tool protection, backend user security, MFA, CSP, QueryBuilder usage, XSS defenses, and CSRF checks. Use when the user asks about TYPO3 security, hardening, permissions, authentication, CSP, SQL injection, XSS, CSRF, or go-live security readiness.
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.