60.5K skills · sorted by relevance
Helps choose and tune Cloud Storage FUSE mounts, which expose bucket objects through file paths. It covers cache sizing, training performance, checkpoint write safety, and cost diagnosis on GKE, Compute Engine, and Cloud Run.
A deprecated guide to building MCP servers with mcp-use, covering callable actions, data resources, reusable prompts, and interactive widgets. It includes design references and code examples, and directs users to the replacement skill, mcp-app-builder.
Helps you write Python code for browser automation with browser-use. Its reference guides cover installation, agent and browser settings, custom tools, model connections, and monitoring so you can configure and extend your automation.
Guides setup for paying for Browser Use Cloud tasks with USDC on Base, a blockchain network. Covers wallet preparation, private key storage, a paid test, and SDK integration, including use without signup or an API key.
Create a new native Muse plugin package with its manifest and requested capability files, then validate it using the installed muse CLI. Receive a file inventory, validation results, and an installation proposal for later use.
Create one new Muse skill with a self-contained SKILL.md and validate it. Build a project skill in the workspace, or prepare a personal skill draft with the command needed to install it.
Measure website performance with Lighthouse, interpret scores and reports, and compare results across builds. Set limits for loading times and resource sizes, then integrate audits into CI, an automated workflow that checks code changes.
Collect posts from X profiles by username, choosing regular posts, posts with replies, or media posts. Receive structured text, author details, engagement counts, and media information for account research and content tracking.
🇺🇸 User Interview & Cold-Start Operations Playbook — Battle-tested framework from HeyGen's 937 interviews to PMF. Complete SOP for user screening, interview execution, Beta testing methodology, cold-start user operations, churn analysis, and user grading systems. 🇨🇳 用户访谈与冷启动运营实战手册 — 基于 HeyGen 937 场访谈验证的 PMF 方法论。包含用户筛选、访谈执行、Beta 测试设计、冷启动用户运营、流失分析、用户分级体系完整 SOP。 🇯🇵 ユーザーインタビュー＆コールドスタート運営プレイブック — HeyGenの937回インタビューで検証されたPMF方法論。ユーザースクリーニング、インタビュー実行、Betaテスト設計、コールドスタートユーザー運営、離脱分析、ユーザーグレーディング。 🇰🇷 사용자 인터뷰 & 콜드 스타트 운영 플레이북 — HeyGen 937회 인터뷰로 검증된 PMF 방법론. 사용자 스크리닝, 인터뷰 실행, Beta 테스트 설계, 콜드 스타트 사용자 운영, 이탈 분석, 사용자 등급 시스템. Triggers: "user interview" | "user research" | "customer discovery" | "PMF" | "product-market fit" | "beta test" | "beta testing" | "cold start users" | "cold start operations" | "churn analysis" | "user grading" | "用户访谈" | "用户调研" | "冷启动" | "冷启动运营" | "Beta测试" | "流失分析" | "用户分级"
The English summary is being prepared. Showing the original description.
Read current or previous Muse Code session logs to recover context, review decisions and tool results, or reconstruct lost files from recorded changes. It also helps locate the relevant session for a workspace.
Move useful memory notes and MCP server settings from Claude Code or Codex into Muse Code. MCP servers connect agents to external tools; the skill checks compatibility and reports settings that need attention.
Delegate a coding task to Muse Code (`muse`) as a background implementer, then review its diff and land it yourself. Use this whenever the user wants to hand implementation work to Muse — phrasings like "have Muse do X", "delegate this to Muse", "run it through Muse", or "use Muse to implement/fix/refactor" — or wants to run a queue of coding tasks through Muse while staying the reviewer. Prefer it when the user will review the diff and commit it themselves. It applies only when the user wants that separate `muse exec` implementer run: if you are Muse yourself and the user simply asks you to write the code, do it directly. DO NOT USE for tasks small enough to do inline.
The English summary is being prepared. Showing the original description.
Diagnose problems in Muse Code itself using the installed executable, settings, crash reports, and session evidence. The skill narrows down the likely failing component, summarizes evidence with sensitive information removed, and recommends a safe next action.
Use this skill to configure and manage NPSP Household Accounts: household naming rules, formal/informal greeting formats, primary contact designation, household merges, and household splits. Trigger keywords: NPSP household naming, household account not updating, merge duplicate households NPSP, household greeting customization, primary affiliation NPSP. NOT for FSC (Financial Services Cloud) households — use admin/household-model-configuration. NOT for bulk-loading constituents and households into NPSP — use data/constituent-data-migration.
The English summary is being prepared. Showing the original description.
Helps you integrate hosted browser automation through Browser Use Cloud APIs and Python or TypeScript SDKs. Reference guides cover authentication, sessions, saved browser profiles, live views, and connections to existing agents or workflow services.
Operate an isolated Browser Use Cloud browser from a machine without usable local Chrome. Navigate pages, inspect content, fill forms, manage tabs, and capture screenshots through the Browser Use CLI, then stop the browser after use.
Publish a saved Muse Idea or turn a completed activity into reusable instructions. The skill prepares a title, summary, and explanation, removes sensitive details, and returns a link for authenticated Muse users.
Turn a talking-head or selfie recording into a vertical video about a Muse experience. Preserve the original narration while showing synchronized chat exchanges, actual work outputs, interface elements, and message sounds to create a shareable clip.
Build MCP servers, which expose tools and data to AI agents, with mcp-use. The material covers server functions, reusable prompts, and interactive React widgets, but this skill is deprecated and points to mcp-app-builder as its replacement.
Connect to and query ClickHouse (a local server or a ClickHouse Cloud service) from the terminal. For ClickHouse Cloud analytics, use the configured direct ClickHouse Query API endpoint with per-user CH_API_KEY and CH_API_SECRET credentials; do not use clickhousectl cloud service query. For local or host/port servers, use clickhousectl local client. Use when the user wants to run SQL against ClickHouse, explore schemas and tables, inspect Cloud services, or authenticate. For building a local dev environment or deploying to Cloud, defer to the official ClickHouse skills (see Scope).
The English summary is being prepared. Showing the original description.
Use this skill to create, deactivate, freeze, or manage Salesforce users, assign user licenses and feature licenses, configure profiles and roles, set login hours and IP restrictions, and set up delegated administration. Triggers: 'add a new user in Salesforce', 'create a user and assign a licence', adding a new user, deactivating a departing employee, license assignment, freezing a user account, delegated admin setup. NOT for designing the permission set / PSG model — use admin/permission-set-architecture. NOT for who can see which records — use admin/sharing-and-visibility. Also covers: bulk user loads by CSV, Bulk API 2.0 upsert on Username, the UserLogin freeze object, licence and permission-set-licence reclamation, Role metadata deployment, and the offboarding checklist as SOQL.
The English summary is being prepared. Showing the original description.
Automate Docuseal operations through Rube MCP, a connection between an AI agent and external tools. The skill searches for supported actions, checks authentication status, and runs requested tasks using the current input requirements for each tool.
Automate Abuselpdb operations through Rube MCP, a connection for agent tools. Discover available actions and their required inputs, verify the service connection, and execute tasks using the current tool definitions.
Resume unfinished Codex work in Muse Code by reading a local session log selected by the user. Recover the latest request, decisions, and pending changes, then check current files and continue the task.