Skills
96,770 packaged capabilities and playbooks for agents. Install any of them with one command.
All
Configures CI/CD pipelines for automated Python package publishing to PyPI or GitLab Package Registry. Use when creating GitHub Actions or GitLab CI release workflows, setting up trusted publishing or API token-based PyPI authentication, configuring version management with git tags and hatch-vcs, writing pyproject.toml publishing metadata, testing packages against TestPyPI, or documenting the release process for a Python project.
Deep learning framework (PyTorch Lightning). Organize PyTorch code into LightningModules, configure Trainers for multi-GPU/TPU, implement data pipelines, callbacks, logging (W&B, TensorBoard), distributed training (DDP, FSDP, DeepSpeed), for scalable neural network training.
IBM quantum computing framework. Use when targeting IBM Quantum hardware, working with Qiskit Runtime for production workloads, or needing IBM optimization tools. Best for IBM hardware execution, quantum error mitigation, and enterprise quantum computing. For Google hardware use cirq; for gradient-based quantum ML use pennylane; for open quantum system simulations use qutip.
Set up or refine open-source .NET code-quality gates for CI: formatting, `.editorconfig`, SDK analyzers, third-party analyzers, coverage, mutation testing, architecture tests, and security scanning. USE FOR: .NET quality gates in CI; analyzer, coverage, mutation, and architecture-test choices; standardizing `.editorconfig`, `dotnet format`, and warning policy. DO NOT USE FOR: non-.NET repositories; generic CI/CD guidance with no .NET quality stack decisions. INVOKES: inspect the repository context, edit targeted files, and run relevant build, test, lint, or validation commands when changes are made.
Guidance for SQL query optimization tasks. This skill should be used when optimizing slow SQL queries, improving database performance, or rewriting queries to be more efficient. Covers query plan analysis, benchmarking strategies, and database-specific optimization techniques.
Build Retrieval-Augmented Generation (RAG) systems for LLM applications with vector databases and semantic search. Use when implementing knowledge-grounded AI, building document Q&A systems, or integrating LLMs with external knowledge bases.
Core React 19 patterns including hooks, Suspense, lazy loading, component structure, TypeScript best practices, and performance optimization. Use when working with React components, hooks, lazy loading, Suspense boundaries, or React-specific TypeScript patterns.
Enforces a disciplined refactoring process for Ruby code: ONE atomic transformation at a time, characterization tests MUST exist and be Green BEFORE any edit, run tests after EVERY single step, rollback immediately on Red (no debugging in broken state), no behavior changes mixed with refactoring. Covers extract methods, rename symbols, split classes, inline variables, remove duplication. Use when: refactor, clean up code, rewrite class, structure changes, simplify.
Safely refactors dbt models with downstream impact analysis. Use when restructuring dbt models for: (1) Task mentions "refactor", "restructure", "extract", "split", "break into", or "reorganize" (2) Extracting CTEs to intermediate models or creating macros (3) Modifying model logic that has downstream consumers (4) Renaming columns, changing types, or reorganizing model dependencies Analyzes all downstream dependencies BEFORE making changes.
Generate release notes in 3 formats (CHANGELOG.md, PR body, Slack announcement) from git commits. Automatically categorizes changes and converts technical language to user-friendly messaging. Use for releases, changelogs, version notes, what's new summaries, or ship announcements.
Reliably show images, renders, mockups, screenshots, or plots to the human user inside the Cursor app chat. Use whenever the user must SEE an image the agent generated or found, or when the user says they cannot see an image you sent. Differentiator: fixes the silent failure where images saved outside the open workspace never render in Cursor chat.
Requirement design workflow - mandatory before any creative development (new features, components, behavior changes, API/DB design): triage, parallel exploration, one-question-at-a-time clarification, adversarial validation and 2-3 option comparison produce a spec, then hand off to writing-plans. Use exploring first while the idea is unsettled; not for pure Q&A, test runs, or no-design-space small fixes (use quick-fix).
Iterative multi-round deep research with structured analysis frameworks. Use for: deep research on a topic, compare X vs Y, landscape analysis, evaluate options for a decision, deep dive into a technology, comprehensive research with cross-referencing. Triggers: deep research, compare, landscape, evaluate, deep dive, comprehensive research, which is better, should we use.
REST API design and development expert specializing in endpoint design, HTTP semantics, versioning, error handling, pagination, and OpenAPI documentation. Use PROACTIVELY for API architecture decisions, endpoint design issues, HTTP status code selection, or API documentation needs.
Optimizes programmers' Markdown resumes: improving technical descriptions, quantifying project results, highlighting technical strengths, and refining the skills section. Use when the user says "optimize my resume", "improve my resume", "rewrite my resume", "make my resume more professional", "improve my resume quality", or provides a resume file to optimize.
Expert in comedic roasts, witty insults, and self-deprecating humor that lands. Covers the craft of punching up vs down, timing, target selection, and knowing the line between funny and mean. Knows how to be savage while staying lovable. Use when "roast, burn, diss, self-deprecating, witty reply, comeback, shade, " mentioned.
Test authenticated routes in the your project using cookie-based authentication. Use this skill when testing API endpoints, validating route functionality, or debugging authentication issues. Includes patterns for using test-auth-route.js and mock authentication.
Imagines the decision has been made and failed — then diagnoses why. Breaks the commitment bias that prevents honest risk assessment after a direction is chosen. Triggers: 'pre-mortem', 'imagine this failed', 'what could go wrong', 'assume this doesn't work', 'failure mode analysis'.
Maps the emotional and political freight that words carry beyond their literal meaning. Use when you say 'this word feels wrong', 'what does this term really signal', 'is this word loaded', 'find a better word for this', 'what are the connotations of', 'why does this language feel off', or when word choice is creating unintended reactions.
Identifies group psychology shaping a discussion or team — groupthink, status dynamics, coalition formation, psychological safety. Triggers: 'group dynamics', 'why does this team behave like this', 'groupthink check', 'why won't people speak up', 'meeting dynamics', 'team dysfunction'.
Diagnoses and repairs dialogue for subtext, voice differentiation, exposition, and forward momentum. Use when dialogue sounds wrong, on-the-nose, or when characters all sound the same. Triggers: 'the dialogue sounds wrong', 'dialogue feels on the nose', 'dialogue review', 'characters sound the same', 'dialogue fix', 'dialogue is too direct', 'the dialogue explains too much'.
Guidance for SAM-based cell segmentation and mask conversion tasks involving MobileSAM, mask-to-polygon conversion, CSV processing, and command-line interface design. This skill applies when working with Segment Anything Model (SAM) for biological image segmentation, converting binary masks to polygon coordinates, processing microscopy data, or building CLI tools that interface with deep learning models. (project)
Guides implementation of EU Standard Contractual Clauses under Commission Decision 2021/914 across all four modules (C2C, C2P, P2P, P2C). Covers clause-by-clause completion, Annex I-III drafting, and SCC module selection. Keywords: SCCs, standard contractual clauses, module selection, data transfers, Annex completion.
Evaluate scientific claims and evidence quality. Use for assessing experimental design validity, identifying biases and confounders, applying evidence grading frameworks (GRADE, Cochrane Risk of Bias), or teaching critical analysis. Best for understanding evidence quality, identifying flaws. For formal peer review writing use peer-review.
Comprehensive toolkit for survival analysis and time-to-event modeling in Python using scikit-survival. Use when working with censored survival data, performing time-to-event analysis, fitting Cox models, Random Survival Forests, Gradient Boosting models, or Survival SVMs, evaluating survival predictions with concordance index or Brier score, handling competing risks, or implementing any survival analysis workflow with the scikit-survival library.
The literature search and citation insertion skill of the SciPilot Skills family. Supports two input modes — Mode A: read the user's Word (.docx) or LaTeX (.tex) paper and insert citations in the corresponding sections of the body text; Mode B: take a passage/an idea/a claim provided by the user and output a list of references that support it (without modifying the document). Both modes search real literature in parallel across three sources — Semantic Scholar, OpenAlex and Crossref — verify authenticity with DOI + cross-validation across sources, then output in the user-specified format (IEEE/APA/Nature/Vancouver/GB-T-7714, etc.). Before every delivery, the Gate 8 end-of-chain hallucination gate audit is run mandatorily; FAIL means delivery is refused. Stage 0 must proactively ask the user about multiple preferences: number of references, format, year range, section, journal quartile, author preferences, institution preferences, whether preprints are included, etc. Silently using default values is forbidden. Trigger this skill when the user mentions any of the following: 添加参考文献, 插入引用, 找文献, citation, reference, 加引用, 补充参考文献, 文献检索, add references, find papers, insert citations, 找文献支撑观点, 用文献支撑这段话, support claim with literature, validate viewpoint with references, or when the user uploads a paper file and asks to add citations, or gives a passage/idea and asks to find literature to support it. It should also trigger even if the user only says "帮我的论文加几篇参考文献" or "scipilot 帮我找文献" or "这个观点有文献支撑吗". Part of the SciPilot Skills research skill family.
Use when the user asks to take a bug, note, defect, improvement, or task from Scout and handle it like a professional AI operator working from Scout. Also use for short commands like "do the next Scout task", "сделай следующую задачу из Скаута", or "возьми задачу из Scout".
Compose semantic vectors in Scry -- embed concepts as @handles, search by cosine distance, debias with vector algebra, and diagnose signal loss. Use when the task involves: semantic search, embedding, vector, cosine distance, <=>, "X but not Y", debias, embed this concept, @handle, vibe algebra, concept vector. NOT for: word2vec training, fine-tuning embeddings, local vector databases (FAISS, Pinecone, Chroma), or plain keyword/SQL search (use scry).
【Internal suite skill · Users usually don’t need to invoke it directly in the full workflow; simply describe the request to the orchestrator short-drama-director, which will route it to this skill automatically】【漫剧老李 V7.4.26 suite · Subskill】Asset-locking specialist engine: extract three lists from the script (character appearances / props / locations) → **produce according to asset form** (M0 appearance profile recap block / **M1 image-generation prompts + image number labels + 《素材接线单》· default** / M2 batch image generation and locking), and manage spatial continuity between locations, single-instance prop locks, and crowd staging. **This package does not hold images**—images are generated by the user on external platforms; this engine must not request images or block downstream work due to missing images. Triggers: break down assets, asset extraction, asset catalog, image-generation prompts, character asset board, location prompts, prop prompts, three lists, 3×3 location image, character portrait, location continuity, prop continuity error, crowd staging, asset form, connection sheet, I have no images. This skill is a suite member; enter the complete workflow through the orchestrator short-drama-director.
【Internal suite skill · Users usually don’t need to invoke it directly in the full workflow; simply describe the request to the orchestrator short-drama-director, which will route it to this skill automatically】【漫剧老李 V7.4.26 suite · Subskill】Action storyboard specialist engine—responsible for generating storyboard prompts for all action, martial arts, and combat scenes: action concepts (motivation / power imbalance / stakes / environmental variables), eight-step writing framework, eight natural-sentence templates for composing each shot, strict-rule self-checks, camera behavior locks, and quick reference for group battles. Triggers: action scenes, martial arts, storyboard fights, combat scenes, ultimate moves, sword formations, magic formations, manifestations, mowing down enemies, exchanges of fire, moves, martial arts, boxing, sword duels, finishers, battlefields, group battles, military formations, camera movement lock. Suite members: orchestrator short-drama-director (authority on system locks and input format) + sd-decision-engine (decisions) + sd-script-engine (scripts) + sd-asset-engine (assets) + sd-storyboard-engine (storyboards) + sd-combat-engine (action scenes · this skill) + sd-feed-engine (feed) + sd-qc-engine (quality control).
【Internal suite skill · Users usually don’t need to invoke it directly in the full workflow; simply describe the request to the orchestrator short-drama-director, which will route it to this skill automatically】【漫剧老李 V7.4.26 suite · Subskill】Director decision engine (basis layer for storyboard selection): scenario mapping E01~E15 + 80/15/5 weighting mechanism + emotional anchor table + composition danger-zone A/B dual-state whitelist; answers “what camera shot should this scene use?” and provides selection criteria for the storyboard engine. Triggers: scenario mapping, what camera angle to use for this scene, emotional anchor, beat sheet, danger zone, composition danger zone, can centered frontal composition be used, what shot should this use. This skill is a suite member; enter the complete workflow through the orchestrator short-drama-director.
【Internal suite skill · Users usually don’t need to invoke it directly in the full workflow; simply describe the request to the orchestrator short-drama-director, which will route it to this skill automatically】【漫剧老李 V7.4.26 suite · Subskill · Track ③ Commentary and narrative】Specialist engine for commentary/narrative videos: complete rule sets for three tracks—historical biographies / historical events and battles / educational explainers. Core capabilities—voice-over script (VO Script) writing standards (commentary speed 4.5–6.5 characters/second / short sentences ≤25 characters / prioritize analogies / reduce cognitive load / hooks and memorable lines), audio-visual information gap discipline, history visual standards prioritizing period settings, hit formulas for three subtypes, storyboard adaptation from narration lines to images, C14 exemption from off-screen voice quotas. Triggers: commentary, shot analysis, roundups, historical figures, historical events, educational explainers, fun facts, educational content, understand at a glance, documentary commentary, talking-head narration. This skill is a suite member; enter the complete workflow through the orchestrator short-drama-director; the track is selected and locked by the orchestrator at 〇·0.
【Internal suite skill · Users usually don’t need to invoke it directly in the full workflow; simply describe the request to the orchestrator short-drama-director, which will route it to this skill automatically】【漫剧老李 V7.4.26 suite · Subskill】Finished-film quality control and continuity management specialist engine: cross-shot / cross-episode consistency ledger, editing transition design and Seedance 15-second segmentation gate, sound design templates, post-production music and sound design (BGM Cue Sheet / music search terms / AI music prompts / copyright checks), and use of automated output checks. **Consistency gates adjust based on asset form** (M0/M1 compare against text profiles; M2 performs image-level verification; this package does not hold images, must not request images or make false verification claims). Triggers: review, quality control, continuity check, watch footage, shot analysis, recheck after dialogue diagnosis, editing transitions, segment checks, sound design, automated checks, add BGM, post-production music, find music, Suno prompts, music copyright, asset form. This skill is a suite member; enter the complete workflow through the orchestrator short-drama-director.
【Internal suite skill · Users usually don’t need to invoke it directly in the full workflow; simply describe the request to the orchestrator short-drama-director, which will route it to this skill automatically】【漫剧老李 V7.4.26 suite · Subskill】Live-action realism combat specialist engine—designs action and produces Seedance 2.5/2.0 feed prompts for modern martial arts, police-and-criminal chases, suspense assassinations, comedy fights, children’s fights, and small-scale group combat (1v2/1v3/2v2/3v3): R1 restrained realism mode, 11-step action dynamics chain, realistic combat technique library (19 martial arts / 36 kicking techniques / 36 grappling techniques), three defensive states and amplified evasions, multi-person combat staging, realistic force feedback, and compliant transcreation. Triggers: live-action fights, realistic combat, modern martial arts, police chases, suspense assassinations, comedy fights, MMA, group combat, one against many, outnumbered combat. Suite members: orchestrator short-drama-director (authority on system locks and input format) + sd-decision-engine (decisions) + sd-script-engine (scripts) + sd-asset-engine (assets) + sd-storyboard-engine (dramatic scene storyboards) + sd-combat-engine (general combat) + sd-xuanhuan-engine (fantasy combat) + sd-real-engine (live-action realism · this skill) + sd-feed-engine (feed) + sd-qc-engine (quality control).
【Suite-internal skill · Users usually do not need to call this directly for the full workflow; simply describe the request to the main controller short-drama-director, which will route to this skill automatically】【Manju Laoli V7.4.26 suite · subskill】Fantasy xianxia combat-specialist engine—creates combat storyboards and feed prompts for fantasy cultivation/xianxia duels, divine and demonic powers, and high-energy Chinese animation: golden combat granularity (2–3-second whole-second cuts; one shot, one move, one response), bidirectional offense-defense power exchanges (preventing one-sided punching-bag fights), four spatial closure protocols for finishing sequences, the four-part scale set and image geometry, 13 three-layer composite elements, a combined magic-and-martial-arts move library, and dual-track physics parameter libraries for realistic and cultivation settings. Triggers: fantasy combat, xianxia duels, spell clashes, Myriad Swords Return to the Origin, sword formations, magical formations, Dharma forms, battles between gods and demons, ultimate finishing moves, xianxia cultivation. Suite members: main controller short-drama-director (authority for locked system and feed format) + sd-decision-engine (decisions) + sd-script-engine (scripts) + sd-asset-engine (assets) + sd-storyboard-engine (dramatic storyboards) + sd-combat-engine (general combat) + sd-xuanhuan-engine (fantasy combat · this skill) + sd-real-engine (live-action realism) + sd-feed-engine (feeding) + sd-qc-engine (quality control).
This skill should be used when the user asks to "search YouTube", "find videos about", "get a transcript", "download subtitles", "extract audio from YouTube", "scan a channel", "research a topic on YouTube", "get video metadata", "what videos exist about", "download YouTube audio", "YouTube research", "summarize this video", "what is this video about", "pull captions from", "grab the audio from", or provides a YouTube/Vimeo/video URL and wants to extract information from it. Also triggers on "batch download transcripts", "analyze a channel", or any multi-video research workflow.
Standardizes security review procedures for Ruby code mapped to OWASP Top 10: allowlist all input params before processing, forbid SQL interpolation (`#{}`), verify no secrets committed or logged, run `bundle exec bundle-audit check --update`, check for shell injection (`system()`, backticks, `exec()`), and discard instruction-like keys (`prompt`, `instructions`) in JSON payloads. Covers input validation, secrets management, and dependency audits. Trigger words: security review, check security, audit code, security vulnerability, secrets check, OWASP.
Comprehensive PET selection guide covering differential privacy, homomorphic encryption, secure multi-party computation, federated learning, zero-knowledge proofs, and trusted execution environments. Includes use-case matching matrix, performance comparison, and GDPR alignment assessment for each technology.
Expert guide for local SEO optimization. Use when optimizing Google Business Profile, managing local citations, NAP consistency, local keyword targeting, review management, local pack ranking, local landing page optimization, or multi-location SEO strategy.
Write a service catalog entry for a microservice or internal platform service — covering service identity, purpose, architecture context, SLAs, API contract summary, data classification, dependencies, operational runbooks, and known limitations. Use when asked to document a service for an internal developer portal, write a service README for a platform catalog, create a service overview page, or onboard a new service to a service registry. Produces a complete service catalog entry suitable for an internal developer portal or wiki.
Master skill — set up the full agent harness for any repo so an agent can work it reliably: legible (map-not-manual docs + custom lints), executable (one-command dev stack), verifiable (e2e gate + a verify-before-ship loop), plus commit hygiene and entropy control. Use when onboarding a new/unfamiliar codebase to agent-driven development — "set up the harness", "make this repo agent-ready", "harness this codebase".
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.