agentsclimarketplace

The catalog

608,119 rows, by stacks then stars

  • Redpen

    sina-z/redpen/.openclaw/skills/redpen Skill

    0★ repo

    Scott red-pens your diff for embarrassing leftovers before you ship: debug output, TODOs, placeholder junk, stray secrets, test artifacts.

  • Automation tool router

    frankxai/starlight-automation-agent-skills/skills/automation-tool-router Skill

    0★ repo

    Choose the best automation layer across native connectors, Codex automations, MCP tools, Make.com scenarios, n8n workflows, and Starlight swarm queues. Use when a user asks which tool to use, how to automate a workflow, whether Make or n8n is better, how MCP should fit, how future agents should remember tool choices, or how to route recurring work safely.

  • Toss invest

    finetension/tosscli-skills/skills/toss-invest Skill

    0★ repo

    Use when a user asks an agent to perform Toss Invest tasks through tosscli, including setup checks, account or asset reads, market data queries, order previews, or order workflows.

  • Managing libraries is a drag

    DovahkiinYuzuko/managing-libraries-is-a-drag Skill

    0★

    Use this skill automatically and without exception whenever ANY new library, package, or dependency that requires third-party license documentation is added to the project, whether explicitly requested by the user or introduced automatically during code modification and package management (e.g., package.json, requirements.txt, Cargo.toml updates).

  • Newbie lens

    dd3ok/newbie-lens Skill

    0★

    Use when reviewing SOPs, PRDs, runbooks, tickets, handoffs, prompts, source-of-truth docs, or attached document bundles for execution-blocking ambiguity so a context-poor human, lightweight model, or agent can follow without guessing.

  • Otf cross platform ui

    otf-kit/skills/skills/otf-cross-platform-ui Skill

    0★ repo

    Build UIs that render the same on web, iOS, and Android from one component API, using OTF's free open-source (MIT) SDK. Use this when starting a cross-platform UI, deciding which OTF package to import, or when you want the same <Button>, <Card>, <Input> — same names, same props, same look — on web and native. Routes to the web (@otfdashkit/ui), native (@otfdashkit/ui-native), and theming (@otfdashkit/tokens) packages.

  • Ganhuo ai affiliate site

    yuanyuanyuan430/ganhuo-ai-growth-skills/skills/ganhuo-ai-affiliate-site Skill

    0★ repo

    Plan and review AI-built affiliate websites for cross-border monetization, including niche selection, reference-site analysis, original positioning, required trust pages, affiliate disclosure, deployability, and review readiness. Use when Codex is asked to build or assess an AI-generated affiliate site, Rakuten application site, or English landing page for affiliate monetization.

  • Skill

    manueldezman/solana-anchor-skill/skill Skill

    0★ repo

    Use this skill for Solana Anchor work including anchor init, scaffold Anchor project, scaffold Anchor app, create Solana Anchor program, install Anchor CLI, anchor command not found, install AVM, fix Anchor setup, write Anchor instruction, review Anchor account, debug Anchor constraint, debug PDA seeds, bump issues, anchor test, test Anchor program, generate Anchor tests, TypeScript Anchor tests, Rust Anchor tests, LiteSVM, solana-program-test, and Surfpool test environments.

  • Video background removal

    lilingm963/removevideobg-skill/skills/video-background-removal Skill

    0★ repo

    Plan and prepare AI video background removal — choose the right tool and export format (transparent WebM/ProRes/PNG, green-screen key, blur, or background replacement), and get ready-to-run ffmpeg recipes for pre- and post-processing (trim, resize, composite an alpha video onto a new background). Use when the user wants to remove or replace a video background, make a transparent video, key out a green screen, or composite footage. The actual AI removal runs at RemoveVideoBG (https://removevideobg.com); this skill advises and prepares the workflow.

  • Daily calendar view

    sidmohan0/skills/daily-calendar-view Skill

    0★ repo

    Produce a read-only daily agenda from connected calendar sources. Use when the user asks to see today's schedule, list calendar events for a day, build a daily calendar view, summarize meetings, identify conflicts/free windows, or gather important context for scheduled events across Google Calendar, Outlook Calendar, and optionally local macOS Calendar.

  • Beach

    ozlar34/packing-agent/packing-agent/skills/beach Skill

    no license0★ repo

    AI agent that builds a weather-aware packing list — merges live forecast, trip-type skills, and a persistent profile (your meds always survive). Google ADK + MCP.

  • Dameng db

    hexinmiao96/dameng-db Skill

    no license0★

    达梦数据库技能包(DM8)。覆盖达梦安装部署、表空间/用户管理、Oracle/MySQL/PostgreSQL/SQL Server/DB2→达梦迁移、SQL/DDL 改写、Java/Spring/MyBatis/MyBatis-Plus/Hibernate 适配、PL/SQL、系统包、性能优化、运维监控和集群架构。 Use when: (1) 安装或配置达梦数据库, (2) 从其他数据库迁移到达梦, (3) 改造达梦兼容 SQL/MyBatis XML/Java 数据源, (4) 处理达梦迁移后的保留字、表别名、GROUP、IDENTITY、自增 ID、字符集、字段大小写等问题, (5) 进行达梦 SQL 开发、性能调优、运维管理或集群方案核对。

  • Goal me

    ecoopnet/goal-me/plugins/goal-me/skills/goal-me Skill

    0★ repo

    Design a goal as a machine-verifiable closed loop so that /goal can run it without drifting or burning excess cost. Interview the user relentlessly to pin down the spec, steps, per-step goals, and a precise stop condition, write them under .claude/goals/, and produce a ready-to-paste kickoff command. Use when the user says things like "set a goal", "goal me", "prepare for /goal", "design a loop", or "I want to build/research X" (but the requirements are vague). Works beyond development (research, writing, investigation) and for both new and existing projects.

  • Rename branch skill

    TopScrech/rename-branch-skill Skill

    no license0★

    Rename the currently checked out Git branch to a requested new branch name both locally and on its remote. Use when the user asks to rename the current branch, selected branch, checked-out branch, or active branch locally and remotely, including updating upstream tracking and deleting the old remote branch after the new remote branch is pushed.

  • Chungus

    oatmealyn-creator/chungus-skills/.opencode/skills/chungus Skill

    0★ repo

    CHUNGUS — Coding God Mode. One install. AI writes minimum code, avoids slop, optimizes React/Next.js, self-heals containers, runs prelaunch audits, follows TDD, and secures your app.

  • Extraction automation architect

    satishTheLegend/extraction-automation-architect Skill

    0★

    Designs and governs a resilient, RECURRING web-scraping / browser-automation system end-to-end — API-first feasibility, a hard legality go-no-go gate (robots.txt, ToS, PII, jurisdiction), backend-agnostic strategy (API > static HTTP > headless browser > managed anti-bot backend, by protection + cost), typed-schema extraction with per-record validation, rate-limiting + backoff, self-healing selectors with change detection, incremental crawling + dedup, structured storage with run history, and monitoring that outlives the native 3-day cron expiry. Backend-agnostic, composes with native /schedule, emits durable artifacts (schema.json, healing.json, monitor.yml, run ledger). Use whenever the user wants a scraper, crawler, or browser automation that keeps working; says it keeps breaking, getting blocked, or returning bad data; wants a site scraped into structured data on a schedule; wants anti-bot, CAPTCHA, rate limits, or proxies handled responsibly; asks if it is legal/ok to scrape something; or wants a scraper monitored with alerts when it breaks — even if they never mention legality, validation, or monitoring and just say "get the data from site X" or "my crawler died". Triggers without "scrape": "pull prices from X", "watch this page for changes".

  • Veeam aiops

    AIops-tools/Veeam-AIops/skills/veeam-aiops Skill

    0★ repo

    Use this skill whenever the user needs to operate Veeam Backup & Replication — a one-shot health overview, read-only diagnostics / RCA (triage failed backup-job sessions and flag repositories low on space), list/inspect/start/stop/retry backup jobs, enable/disable jobs, list restore points and start a VM restore, list backup repositories with capacity, list stored backups and their objects, inventory backup infrastructure (managed servers, proxies), and poll/stop async sessions for job/restore progress. Always use this skill for "list veeam jobs", "run veeam backup", "start veeam job", "veeam restore", "veeam repository", "veeam backup status", or "veeam session" when the context is explicitly Veeam / Veeam Backup & Replication / VBR. Do NOT use when the target is not Veeam Backup & Replication (other backup products, hypervisor lifecycle, or cloud providers are out of scope). Common Veeam B&R operations with a built-in governance harness (audit, policy, token budget, undo, risk-tiers).

  • Docx

    NyXXiR/skillboard/examples/multi-source-skills/anthropic/docx Skill

    0★ repo

    Create and edit Word documents.

  • Code review

    draculavan/agent-skill-atlas/examples/demo-skills/code-review Skill

    no license0★ repo

    Review source changes for risk, regressions, missing validation, and maintainability.

  • Orz slides skills

    wangyu16/orz-slides/orz-slides-skills Skill

    0★ repo

    Author and edit single-file, self-editing .slides.html presentation decks (orz-slides). Use when a user wants to turn notes or markdown into a browser-based slide deck — one portable HTML file that presents in any browser (reveal.js), is written in orz-markdown with a layout syntax, and can be edited in-browser. Covers the deck source format, slide/layout/region markers, structure templates, and per-slide capacity budgets so slides do not overflow.

    The same file is in 1 other repository

  • Perf optimization loop

    satishTheLegend/perf-optimization-loop Skill

    0★

    Runs a disciplined, profile-gated performance optimization campaign that REFUSES to change code without a measured baseline, keeps a change only when a re-measured delta beats noise (else auto-reverts), records every hypothesis — wins AND reverted negatives — in a persistent cross-session optimization ledger, and locks each win behind a generated CI performance budget so it can never silently regress. Works across web frontends (LCP/INP/CLS), Node/Python/Go/JVM backends (CPU/memory/IO/allocations), and database query hotspots (EXPLAIN/N+1/indexes) in one loop. Use whenever the user says something is slow — "this page is slow", "the endpoint has high latency", "why is this so slow", "profile this", "find the bottleneck", "fix Core Web Vitals", "reduce LCP/INP", "this query takes seconds / N+1", "our compute bill is high", or "stop performance regressions / add a perf budget" — even if they don't ask for a process, a profiler, or a benchmark, and even if they just paste a slow function, a flamegraph, a query plan, or a Lighthouse report. Trigger even when the user only complains about speed or asks you to "make it faster" without mentioning measurement.

  • Jgs reference skill

    jgsystemsconsulting/jgs-reference-skill Skill

    no license0★

    Converts an authoritative reference document (standard, handbook, guidebook, framework: PDF/EPUB/DOCX/HTML/MD/RTF) into a licence-clean, citable knowledge pack: a progressive-disclosure agent skill with SKILL.md + chapters + glossary + patterns + cheatsheet, plus PACK.yaml provenance and a per-pack LICENSE. Use when you want a trustworthy reference oracle over a vetted open source (not study notes). Vets the source's licence FIRST and refuses to package non-redistributable (paywalled/all-rights-reserved) sources, routing them to a citation-only signpost instead.

  • Orchestra memory

    luigipascal/skill-forge/.cursor/skills/orchestra-memory Skill

    no license0★ repo

    Queries Orchestra shared memory at localhost:4317 before coding when context may exist in prior chats, synced repos, or ingested threads. Use when the user mentions Orchestra memory, Untitled threads, recall, or cross-project context.

  • Aisp creator evolution aisp

    AIXP-Labs/SoulSkill/claude_code_plugin/skills/aisp_creator_evolution_aisp Skill

    0★ repo

    AISP-backed bridge for Create new conformant AISP skills, evolve existing ones, and adapt existing Agent Skills into native AISP packages from requirements, SOPs, prompts, or evidence. Use when create a new AISP skill; convert a requirement, workflow, SOP, or prompt into an AISP skill. Do not use when research only, with no skill produced; generating a non-AISP artifact (e.g. a plain SKILL.md with no AISOP body).

    The same file is in 1 other repository

  • Project context pack

    FairladyZ625/project-context-pack/skills/project-context-pack Skill

    0★ repo

    把一个模糊的项目目标,通过和用户**一个问题一个问题、无限循环地持续讨论**(对抗式追问、反思、主动澄清),逐渐逼到清晰,并渐进沉淀成一套分层、自包含、任何人读完即可开工的设计上下文包。当用户描述了一个还不完全清晰的目标——"我想做一个 X""我们重写 / 重新设计 / 改造一下 Y""帮我想清楚这个东西到底要做什么"——务必使用本 skill。它适用于全新项目和老项目改造 alike。即使用户没明说,只要目标还模糊、需要持续讨论才能清晰,就主动触发它,不要等用户说"我要做设计文档"。

  • Laravel eloquent discipline

    shaxzodbek-uzb/laravel-guardrails/skills/laravel-eloquent-discipline Skill

    0★ repo

    This skill should be used when the agent writes or edits Eloquent models, controllers, actions, jobs, or repositories — specifically when it calls create(), update(), fill(), forceFill(), all(), get(), whereRaw, orderByRaw, selectRaw, DB::raw, or mass update()/delete(); when it touches $fillable, $guarded, $casts, $appends, unguard(), updateOrCreate, firstOrCreate, or upsert; or when it passes request()->all()/$request->all() into a model. Also load when reviewing migrations-to-model mapping, dealing with money/decimal/boolean/date/enum/JSON columns, exporting or iterating large tables, or wiring user-supplied sort/filter parameters into a query. Keywords: mass assignment, MassAssignmentException, SQL injection, casts, fillable, guarded, unbounded query, memory exhaustion, chunk, cursor, lazy, paginate.

  • Skill

    mitgajera/solana-program-upgrade-skill/skill Skill

    0★ repo

    Safely upgrade live Solana programs and migrate on-chain account data without bricking the program or corrupting user accounts. Use whenever the user changes an `#[account]` struct (adds/removes/reorders/retypes fields), redeploys or upgrades a deployed program, hits a size-mismatch or `AccountDidNotDeserialize` error after an upgrade, works with `realloc`/buffer accounts/`solana program extend`, manages upgrade authority or a Squads multisig, makes a program `--final`/immutable, or plans, simulates, or rolls back a program upgrade - even if they never say the word "migration." Trigger on any program upgrade, redeploy, account layout change, schema change, or upgrade-authority topic.

  • Company news

    yogeshg665/quill-trading-agent/skills/company-news Skill

    0★ repo

    Turn aggregated daily company-news sentiment into a small, deterministic directional trade proposal. WHEN: "trade on this news", "act on the headline", "buy on positive news", "trim on negative news", "news-driven trade", "sentiment-based proposal".

  • Cx sentiment analysis

    admaduranga/agent-council-cx-sentiment-skill/skills/cx-sentiment-analysis Skill

    0★ repo

    Analyzes SaaS technical support interactions using a six-dimension sentiment and churn-risk rubric with deterministic precedence scoring, verbatim evidence quotes, and strict JSON output. Use when scoring support tickets, evaluating customer sentiment, assessing churn or escalation risk, CX QA, or when the user mentions support case sentiment analysis.

  • Add feature design

    falconluca/skills/skills/add-feature-design Skill

    no license0★ repo

    用设计驱动的方式给现有代码库添加新功能,并在当前项目的 docs/ 下先输出计划文档,等待确认后再实现。适用于用户要求实现新功能、扩展已有业务能力、接入新用例、增加新策略或新流程,并要求代码具备第一性原理分析、严格面向对象设计、类模型拆分、目录结构抽象、模块/子目录分层、对象协作编排、可扩展性、可测试性、低耦合、关注点分离、依赖倒置、依赖注入、IoC、状态建模、Mermaid 图示、结构化日志规约、减少魔法值、减少分支复杂度、避免代码臭味或边做边重构的场景。

  • Failsafe readiness

    e3742526/agent-skills-for-quality/skills/failsafe-readiness Skill

    0★ repo

    Audit whether software fails safely under missing dependencies, bad configuration, malformed input, stalled work, interrupted runs, network errors, partial data, and operator mistakes. Use when reviewing graceful startup, shutdown, cleanup, timeout, recovery, degraded mode, logging, and rerun safety.

  • Zhuoxiang skill

    sdyuyouth/zhuoxiang-skill Skill

    0★

    防止"着相"——死抠用户字面表述、忽略真实意图。在三类场景下使用:用户给出需求或具体方案时(先反推目标,识别 XY 问题);用户对输出提出修改反馈时(判断是个案还是通用原则,主动举一反三);用户提问内嵌前提断言时(先验证前提再动手)。适用于任何"根据用户输入做回复、方案或修改"的任务。

  • Device it

    Christian-Katzmann/device-it Skill

    0★

    Put a local web project on a device Home Screen / launcher as a launchable, fullscreen, offline-capable app with its own icon and name — iPhone, iPad, Android, or desktop browser. Use when the user says device-it, "put this on my phone/iPad/tablet", "make this an installable app", "install on my device", "home screen app", or wants the on-device sibling of app-it (which makes a Mac Dock app). Builds the project, applies a PWA transform (manifest, icons, service worker) to the built output only, deploys to a stable HTTPS URL, and installs via the pocket MDM on Apple devices (zero-touch: icon just appears) or a QR/Add-to-Home-Screen link everywhere. Also handles update, remove, list, doctor, and one-time setup/enrollment. Legacy alias: ipad-it.

  • Prototype iteration protocol

    lhw0414/prototype-iteration-protocol Skill

    0★

    Vibe coding 迭代已有原型/demo 时的协作协议。适用于本地可编辑、可重复运行或预览的 demo (网页、原生 App、脚本等),不限操作系统(macOS / Windows / 鸿蒙等)。 Use when: 用户要在已有 demo 上修改——如「改一下 X」「改一下 X 和 Y」「调整某交互」「把某处改成…」; 或说「只改」「别动其他的」「存一版」「做一版新的」「回到某版」;或 demo 越改越乱、改一处坏别处。 不用于:从零新建项目、与当前 demo 无关的问答。 Triggers: 改一下、改下、调整、改成、修改、再改、迭代、存一版、做一版新的、回到某版、 只改、别动、 change, tweak, adjust, modify, iterate, save version, new version, rollback

  • Codebase map

    peterphoenix/deep-reasoning/skills/codebase-map Skill

    0★ repo

    Generate and maintain persistent project knowledge (an agent memory file such as AGENTS.md or CLAUDE.md, plus ARCHITECTURE.md) so the agent starts every session already knowing the codebase. Use when: starting work in a repository that has no memory file or a stale one; the user says 'onboard', 'index this repo', 'remember this project', 'set up project memory', or 'update project context'; or after completing significant work that changed the architecture, conventions, or gotchas worth persisting. Also use proactively — if a session begins in a repo without a memory file, offer to generate one.

  • Frontend council

    FantaStudio/frontend-council Skill

    0★

    Your frontend second opinion: three developers debate, you get the verdict. Junior, middle, and senior perspectives from different corporate projects. Use for "call the council", "frontend-council", contentious frontend decisions. Works in English and Russian.

  • Gpt engineer spark

    SYMBaiEX/gpt-orchestration/skills/gpt-engineer-spark Skill

    0★ repo

    Lead end-to-end engineering work with a capable main agent and a parallel fleet of model-pinned GPT-5.3-Codex-Spark subagents. Use when the user asks for a Spark fleet, ultra-fast parallel coding agents, rapid repository exploration, many bounded implementation shards, or low-latency independent verification while retaining architecture, integration, and final acceptance with the main agent.

    The same file is in 1 other repository

  • Explain

    danjutan/explain/skills/explain Skill

    0★ repo

    Build a concept-ladder explainer as a self-contained HTML page, in a background agent, and hand off a working address for it.

  • Document skills windows

    NobuoTsukamoto/claude-document-skills-setup/skills/document-skills-windows Skill

    no license0★ repo

    Windows で公式 document-skills (docx / pptx / xlsx) を使って Word / PowerPoint / Excel ファイルを作成・編集・PDF化・検証するときに、公式スキルと必ず併用する Windows 補完スキル。Unix 専用スクリプトの回避策・専用 venv の Python パス・xlsx 再計算の代替手順を提供する。Use alongside the official docx/pptx/xlsx skills whenever creating, editing, converting, or validating Word, PowerPoint, or Excel files on Windows.

  • Error as value

    spotsccc/error-as-value/skills/error-as-value Skill

    0★ repo

    Apply the errors-as-values convention in TypeScript repositories that use @spotsccc/error-as-value or Error | T return unions. Use when implementing or reviewing typed errors, converting throwing dependencies at boundaries, adapting value errors to exception-based consumers, or handling cancellation and cause chains. Prefer the smallest useful error model. Do not introduce this convention where another error model is established unless the user requests a migration.

  • Fundus

    grischaerbe/fundus-skill/skills/fundus Skill

    no license0★ repo

    Use Fundus to import, replace, reimport, organize, process, validate, preload, and render production assets in Svelte 5 mobile app projects that already declare Fundus, contain fundus.config.ts, or are explicitly adopting Fundus. Use for Fundus setup, asset libraries, local files, Figma selection sources, images, slices, video, chroma-key video, audio, manifests, generated modules, the Fundus CLI or editor, runtime components, delivery budgets, and preload lifecycles.

  • Api layer boundaries

    Sarmkadan/dotnet-senior-skills/skills/api-layer-boundaries Skill

    0★ repo

    Enforce layering in ASP.NET Core services - what belongs in controllers, application services, and repositories; DTO vs entity leakage; where transactions and validation live. Use when reviewing or designing API endpoints and service classes.

  • Mac migration operator skill

    stephanechery/mac-migration-operator-skill Skill

    0★

    Use this skill when a user is planning, running, validating, or recovering a Mac-to-Mac migration. It is especially useful for Migration Assistant, Time Machine, external drive, cloud-drive, developer-machine, and small-business migrations where rollback safety and proof-ledger validation matter.

  • Readme bilingual

    nicaijimi/readme-bilingual Skill

    no license0★

    Generate a high-quality bilingual (Chinese/English) README from a single-language source. Not just translation — adapts idioms, explains culture-specific references, and follows technical writing conventions in both languages. Trigger when the user asks to create a Chinese or English version of their README, docs, or any technical text.

  • Answer shapes

    Tsurai7/agentic-playbook/skills/answer-shapes Skill

    0★ repo

    Optional output skeletons for common coding deliverables — bug fix, code review, investigation. Apply when the deliverable is one of these shapes to keep the answer complete and parseable (root cause / fix / verification, etc.); especially helps weaker models and machine-consumed handoffs. Skip for freeform questions.

  • Regitize authors

    gadost/regitize/skills/regitize-authors Skill

    0★ repo

    Normalize historical Git author names and email addresses from repository .mailmap rules, preview every identity change, and safely rewrite one local branch after explicit approval. Use when the user invokes regitize-authors, asks to apply .mailmap permanently to commit history, wants duplicate author identities consolidated, or needs historical author metadata corrected.

  • Audit docs

    LandonTomaine/LT-Agents/skills/audit-docs Skill

    0★ repo

    Lightweight documentation audit workflow for checking repo docs, indexes, and agent guidance for stale content, duplication, missing routes, unclear ownership, and mismatches with the current repository. Use when the user asks to review, tighten, de-duplicate, or improve documentation without doing a broad standards bootstrap.

  • Design taste frontend

    skill-desktop-labs/skill-desktop/.agents/skills/design-taste-frontend Skill

    0★ repo

    Anti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-flight check.

    The same file is in 16 other repositories

  • Recall

    themarchrain/mematrix/skills/recall Skill

    no license0★ repo

    Load mematrix memories into the current session context. Use when user invokes /mematrix:recall, /mematrix:read, or asks to "load my memories", "recall context", "what do you remember about me", "read my profile", or needs background context from previous sessions.

  • Goal draft policy

    yshmr/claude-code-goal-draft-policy/goal-draft-policy Skill

    0★ repo

    Author, critique, or repair a Claude Code `/goal` completion condition — the autonomous loop where Claude keeps working across turns until a condition is met. Use this whenever the user wants to set a goal, "let Claude run until X is done", write or fix a `/goal` condition, run a long autonomous task with a verifiable end state, or asks why their goal isn't completing (or is completing too early). Trigger even if the user doesn't say the word "goal" — e.g. "keep going until all tests pass without me approving each step", "make Claude finish this migration on its own", "have it loop until the build is clean". Produces a ready-to-paste `/goal ...` line, not prose.

  • Antigen epitope pipeline

    ruthannepai-tech/inflammatory-target-discovery-skills/skills/antigen-epitope-pipeline Skill

    0★ repo

    Map antigen proteins to MHC-II/I binding epitopes and per-HLA burden for a personalized, disease-agnostic pMHC pipeline. Use when building food/auto/microbial-antigen epitope panels (EoE, celiac, T1D, IBD), predicting which peptides a patient's HLA can present, or designing antigen-specific T-cell diagnostics/tolerance reagents. Covers antigen->UniProt->peptide-window->MHC-prediction->burden, and the presentation-vs-pathology guardrail.

  • Copy court

    AliceEinstein/copy-court Skill

    0★

    A merciless editor that scans any piece of writing for formula, template, AI slop, borrowed brand voice, self-repetition, and hollow copy. Works on blog posts, social posts, sales decks, website and landing copy, ads, email, video scripts, and resumes. Use this skill whenever someone asks you to review a draft, sanity-check copy, tell them if something reads generic or "sounds like AI," give a brutally honest second opinion, or catch cliches and templated structure. Also reach for it as the final gate before anything gets published. Trigger it even when the user does not say the words "template" or "review," as long as they are handing you writing and want to know what is wrong with it.

  • Multi perspective delivery

    kangkang07890-gif/multi-perspective-delivery/skills/multi-perspective-delivery Skill

    0★ repo

    Deliver or review a concrete product, AI-agent, architecture, or software artifact or decision through distinct perspectives, then synthesize one evidence-backed result. Use when the user explicitly asks for multi-perspective, cross-functional, multi-role, expert-panel review of a concrete artifact or decision, product-engineering-QA, or independent acceptance review; or when a high-risk architecture, security, data, production, or migration decision genuinely requires different disciplines. Do not trigger for open-ended persona roundtables, factual answers, minor edits, routine documentation, isolated code questions, or ordinary implementation and testing unless multiple perspectives are requested.

  • Repo visibility skill

    josezuma/repo-visibility-skill Skill

    0★

    Audits a GitHub repository for AI discoverability. Checks AGENTS.md presence, README structure, topics, description quality, and llms.txt. Use when evaluating how discoverable a repo is to AI coding agents and AI search.

  • React prototype system

    1012081579/react-prototype-system/react-prototype-system Skill

    0★ repo

    Turn Figma frames, screenshots, visual specifications, or existing UI designs into high-fidelity, performance-aware React prototypes with Figma-aligned naming, Tailwind-backed design tokens, content-verified assets, responsive named component files, and a component-driven workflow. Default new or unconstrained implementations to TypeScript and Tailwind CSS. Use when Codex is asked to reconstruct React components, screens, flows, design systems, product concepts, or front-end mock applications from a design source; evolve a prototype without collapsing its architecture; or review design-to-code fidelity, asset integrity, naming, token use, component boundaries, async waterfalls, bundle boundaries, server/client data transfer, re-renders, interaction states, responsiveness, accessibility, and composition.

  • Hello

    connordibble/dibble/examples/marketplace-kit/broken-marketplace/plugins/greeter/.claude-plugin/skills/hello Skill

    0★ repo

    Greets the user

  • Ui critique coach

    linxumoney/ui-critique-coach/skills/ui-critique-coach Skill

    0★ repo

    评审网页、App、后台系统、表单、原型或截图的用户任务、信息层级、交互状态、可访问性、响应式行为和视觉一致性,输出按 P0-P3 严重度排序的证据化修改建议。适用于 UI/UX 评审、页面改版、截图分析、设计验收、移动端检查和“这个界面哪里有问题”。

  • Drag delete

    treramey/drag/skills/drag-delete Skill

    0★ repo

    Delete Tempo Cloud worklogs with Drag. Use when the user explicitly asks to preview or delete one or more worklogs by numeric ID.

  • Build

    bingelp/skills/skills/build Skill

    no license0★ repo

    Implements tasks from an approved plan, one at a time. Only runs when the user explicitly types /build.

  • Poster generator

    nanoodlecom/nanoodle-js/examples/agent-skill/poster-generator Skill

    0★ repo

    Generate a poster image from a short idea — an LLM expands the idea into a detailed image prompt, then an image model renders it. Use when the user asks for a poster, illustration, or promo image from a one-line concept.