发现 Skills

OpenClaw plugin for documentation-grade memory: explicit capture + local searchable store with safe redaction.

Extract frames or short clips from videos using ffmpeg.

Explore options and constraints before planning without implementing code.

Download videos from YouTube, Bilibili, Twitter, and thousands of other sites using yt-dlp. Use when the user provides a video or playlist URL and wants to download the file, extract audio (MP3), download subtitles, or choose video quality. Triggers on phrases like download video, yt-dlp, YouTube, Bilibili, Douyin, TikTok, extract audio.

Google Model Armor: Filter user-generated content for safety.

Creates unit test YAML definitions that mock upstream model inputs and validate expected outputs. Use when adding unit tests for a dbt model or practicing test-driven development (TDD) in dbt.

Handle tool failures when researching regulatory/government content by using fallback methods and domain knowledge

This skill should be used when the user asks about Polars DataFrame library (Apache Arrow) for Python or Rust. Triggers: "polars expressions", "lazy vs eager", "scan_parquet streaming", "convert pandas to polars", "pyspark to polars", "kdb to polars", "group_by_dynamic", "rolling_mean", "polars window functions", "asof join", "polars GPU", "polars parquet", "LazyFrame". Time series: OHLCV resampling, rolling windows, financial data patterns. Performance: native expressions over map_elements, early projection, categorical types, streaming.

Manage user expectations during wait times with appropriate loading states — from simple spinners to complex skeleton screens and staggered animations. Perceived performance is often more important than actual load time. Use when designing data-heavy components, handling API calls, building hero sections, or improving the feel of a slow interface.

Mao Zedong’s thinking framework and modes of expression. Based on deep research of 100+ primary and secondary sources, this distills from Selected Works, collected writings, poetry, dialogue records, decision archives, and external studies: seven core mental models, ten decision heuristics, and a complete expressive DNA. Purpose: act as a thinking advisor that analyzes strategic problems, reviews decisions, and assesses situations from Mao Zedong’s perspective — using contradiction analysis, seeking truth from facts, the mass line, a philosophy of struggle, strategic-level disdain with tactical emphasis, independence and self-reliance, and Marxism adapted to China. Use when the user expresses keywords or intents such as: “from Mao Zedong’s perspective,” “how would Mao see this,” “Mao-style thinking,” “contradiction analysis,” “from the perspective of ‘seek truth from facts’,” “class analysis,” “mass line,” “instructor perspective,” “use the thinking of the Selected Works,” “analyze using On Practice/On Contradiction,” or “view from a class angle.” Even brief prompts like “help me think with Mao’s methods,” “what would an instructor do,” “analyze dialectically,” “switch to a worker–peasant perspective,” “help me identify the principal contradiction here,” or “from this angle, who are our enemies and who are our friends” should trigger it. When not to activate: do not activate when the user is only discussing Mao’s biography or historical evaluation (and is not asking for analysis from his perspective); if the user only says “analyze dialectically” without any Mao/Selected Works context, prefer generic dialectical reasoning rather than this skill.

Pirate metrics — Acquisition, Activation, Retention, Referral, Revenue — adapted for B2B

A/B test design — primary metric, MDE, sample size, duration, guardrails, critical region

Turning data into narratives — data story types, persuasion techniques, data-driven visualization, platform adaptation

Find and analyze provided design materials (files/links), determine the "source of truth" for UI, and extract checkable rules (screens, components, tokens, states).

Epic decomposition into stories with dependencies, size estimates (S/M/L/XL), critical path

Generating visual content concepts using Google Stitch (AI) with MCP — carousels, cover images, banners, social media visuals

JTBD canvas for B2B products — functional, emotional, social jobs separately for buyer and end-user

Objectives & Key Results for B2B product teams — quarterly cycle, measurable key results

Product problem formulation — actor, context, pain, root cause, evidence

Thematic roadmap — Now / Next / Later, without fixed dates, attached to OKRs

name: react-beast-practices description: Senior-level React best practices: composition, container/presenter, custom hooks, Context, performance (React.memo/useMemo/useCallback), code splitting, error boundaries

B2B customer-facing release notes — what's new, why it matters, breaking changes, migration steps

Expert in backend development. Provides specialist support in API design, server-side logic, authentication, authorization, database integration and third-party service connections. Works with Node.js, Express, FastAPI, NestJS, Django and Python. Use this agent when the user mentions concepts like API, endpoint, backend, server, auth, JWT, middleware, REST, GraphQL, webhook, cron job, queue, etc.

Elite frontend architect who treats UX as art, state as design, and performance as engineering. Builds scalable 60fps experiences with elegant, maintainable architectures. Use for state management, frontend architecture, performance optimization, Web APIs, optimistic UI, or progressive enhancement.