AI & ML
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
-
knownasnaffy Bundle Save Token 5💰 Save Token | Token 节省器 TRIGGERS: Use when token cost is high, conversation is long, files read multiple times, or before complex tasks. Guiding skill that helps agents identify and avoid sending duplicate context to LLM APIs. Teaches agents to recognize repeated content and summarize instead of re-sending. 触发条件:Token 成本高、对话长、文件多次读取、复杂任务前。 指导 Agent 识别重复内容,避免重复发送,从而节省 Token。
0 -
knownasnaffy Bundle Polymarket 10Query Polymarket prediction market odds and events via CLI. Search for markets, get current prices, list events by category. Supports sports betting (NFL, NBA, soccer/EPL, Champions League), politics, crypto, elections, geopolitics. Real money markets = more accurate than polls. No API key required. Use when asked about odds, probabilities, predictions, or "what are the chances of X".
0 -
knownasnaffy Bundle Polymarket Candle Marubozu Trader 7Trades marubozu continuation signals on Polymarket 5-minute crypto interval markets. A marubozu is an interval with extreme conviction (>65% or <35%) indicating full-body directional commitment with no hesitation. The next interval should continue in the same direction because extreme conviction reflects new information that has not yet fully propagated. Targets BTC, ETH, SOL, and XRP Up or Down bundles with conviction-based position sizing.
0 -
knownasnaffy Bundle Svg Animator 2Generate animated videos from SVG frames using text LLM. Supports any subject (animals, humans, characters, scenes, abstract art), automatic duration calculation, and multi-scene story animations. Perfect替代 AI 视频生成 API,直接用文本模型生成 SVG 代码然后合成视频。
0 -
knownasnaffy Bundle Context Guardian 5Production-ready context continuity skill for autonomous AI agents. Use when tasks may outlive a single LLM context window, when you need durable checkpoints, structured summaries, restart recovery, pressure-aware trimming, or when the agent must stop safely instead of continuing blindly after context loss.
0 -
knownasnaffy Bundle Document Learning 6Comprehensive document learning system with progress tracking, resume capability, and long-term memory integration. Use when you need to read PDF/text documents, track your learning progress (chapter/page bookmarks), resume from where you left off in future sessions, and automatically store learned knowledge into MEMORY.md for permanent retention. Supports large file handling and chunked processing.
0 -
knownasnaffy Bundle Geojson 6Inspect, validate, summarize, and troubleshoot GeoJSON files and payloads, including FeatureCollection checks, geometry counting, bbox generation, coordinate range review, and CRS-related safety guidance. Use when the user asks about GeoJSON, FeatureCollection, geometry validity, 坐标范围检查, bbox, GeoJSON summary, or GeoJSON debugging.
0 -
knownasnaffy Bundle Feishu Audio 3Send TTS audio as a proper playable audio message (not file attachment) to Feishu chats. Use when asked to send voice messages, TTS audio, speech announcements, or 语音报数 (voice roll calls) to a Feishu group or user. Activates on requests like "send voice message to Feishu", "TTS to Feishu chat", "语音发送", "语音报数", or any task combining text-to-speech with Feishu messaging.
0 -
knownasnaffy Bundle Polymarket Edge Liquidity 4Trades active Polymarket-imported markets on Simmer when estimated edge and liquidity filters pass.
0 -
knownasnaffy Bundle Git Workflow 5Use this skill for any git commit, pull request, or release task. Invoke immediately when the user wants to: stage and commit changes, write a commit message, push code, create a PR or MR, tag a release, or update CHANGELOG. Triggers on: 「帮我提交」「commit」「提交代码」「创建 PR」「发布版本」「打 tag」 「推代码」, or English equivalents like "commit my changes", "create a pull request", "release v", "write a commit message", "push this branch". Do NOT trigger for: bulk repo syncing (use code-sync), general git questions without commit/PR intent, or Yunxiao MR creation (use yunxiao skill).
0 -
knownasnaffy Bundle JWT Toolkit 2Decode, inspect, and validate JWT (JSON Web Token) tokens from the command line. Shows header, payload, algorithm, expiry status, and known claim labels. Use when debugging auth tokens, checking if a JWT is expired, inspecting JWT claims, decoding Bearer tokens, or analyzing token structure. Triggers on "decode JWT", "inspect token", "JWT expired", "parse JWT", "check Bearer token", "token claims".
0 -
knownasnaffy Bundle Visa Itinerary Gen 7一键生成领馆级签证行程计划书 — Generate consulate-grade visa itinerary from natural language. Real flyai data, zero hallucination. PDF + booking links with Fliggy.
0 -
knownasnaffy Bundle N8n Builder 2Expert n8n workflow builder that creates, deploys, and manages n8n workflows programmatically via the n8n REST API. Use when asked to create n8n workflows, automate n8n tasks, build automations, design workflow pipelines, connect services via n8n, or manage existing n8n workflows. Handles webhook flows, scheduled tasks, AI agents, database syncs, conditional logic, error handling, and any n8n node configuration.
0 -
knownasnaffy Bundle Openclaw Skill Creator 3Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md file. Also use when editing or restructuring a skill directory (moving files to references/ or scripts/, removing stale content, validating against the AgentSkills spec). Triggers on phrases like "create a skill", "author a skill", "tidy up a skill", "improve this skill", "review the skill", "clean up the skill", "audit the skill".
0 -
knownasnaffy Skill Background Music Video 2Background Music Video - Add Background Music to Any Video with AI Chat. Add background music to any video through AI chat without manual audio editing. Upload a video and describe the mood you want: "add upbeat music for a travel vlog" or "put calm piano background for a cooking tutorial" or "match energetic music to this workout montage." The AI selects fitting tracks from a built-in royalty-free library, adjusts volume levels, and syncs the music to your video's pacing. Handles automatic music selection based on video mood and content, volume ducking when speech is detected, seamless music looping for longer videos, fade-in and fade-out at video start and end, multiple music tracks layered for different sections, and custom volume adjustment through chat commands. No audio editing software, no waveform scrubbing, no manual sync. Built for YouTube creators adding background music without copyright strikes, podcast editors adding intro and outro music, social media managers scoring content for Instagram and
0 -
knownasnaffy Bundle Veillabs 3Interact with Veillabs privacy-focused DEX API for cross-chain swaps, private seed distributions, and transaction tracking.
0 -
knownasnaffy Bundle Yield 4Conversational Compounding Engine — models every bot message as a financial investment that compounds trust, micro-commitments, and conversion momentum over time. Zero external APIs. Zero cost. Maximum revenue lift.
0 -
knownasnaffy Bundle App Store Changelog 6Create user-facing App Store release notes by collecting and summarizing all user-impacting changes since the last git tag (or a specified ref). Use when asked to generate a comprehensive release changelog, App Store "What's New" text, or release notes based on git history or tags.
0 -
knownasnaffy Bundle Perfetto Analyse 5Captures and analyzes Android Perfetto traces via CLI and config files. Covers trace collection (CPU, GPU, memory, rendering, power), editing TraceConfig (.pbtx), and analyzing jank, latency, overdraw, and power. Trace query can be done in Perfetto UI or via trace_processor command-line tool (PerfettoSQL). Use when working with Perfetto, systrace, Android performance, trace configs, or analyzing .perfetto-trace files.
0 -
knownasnaffy Bundle Failure Monitor 5Automated failure detection, diagnosis, and auto-repair system for cron jobs. Monitors task health, fixes common issues automatically, and requests human confirmation when needed.
0 -
knownasnaffy Bundle Meeting Notes Tasks Follow Up 4Lightweight meeting-notes productivity workflow for turning transcripts or raw meeting notes into clean summaries, extracted key points, and simple action lists. Use when processing meeting transcripts, call notes, workshop notes, or internal sync notes into a clear recap and a basic task list for a free ClawHub edition.
0 -
knownasnaffy Bundle Catl Wiki 5CATL (宁德时代) project wiki knowledge base — shared across all agents. Activate when: (1) looking up CATL client info, project docs, industry research, content assets, or SOPs, (2) adding/updating wiki content for the CATL project, (3) user mentions 宁德wiki, 知识库, client profile, 项目文档, 行业研究, or any CATL project knowledge query. Provides wiki node mapping, read/write workflows, and mandatory changelog tracking. NOT for: non-CATL wikis, general feishu doc editing unrelated to the knowledge base, or feishu drive/permission management.
0 -
knownasnaffy Bundle Notebooklm PDF Cleaner 3Create a presentation-ready copy of a NotebookLM-exported slide-deck PDF by masking the small visible NotebookLM footer badge at the bottom-right of each page. Use when the user wants a cleaner NotebookLM PDF export for sharing, presenting, or emailing. This skill is narrowly scoped to NotebookLM slide-deck PDFs and does not act as a general watermark-removal tool.
0 -
knownasnaffy Bundle Scrapling Web Scraping 4Zero-bot-detection web scraping for OpenClaw. Bypass Cloudflare, handle JavaScript-heavy sites, and adapt to website changes automatically. Use when you need to scrape protected websites, extract data from dynamic JavaScript SPAs, or bypass anti-bot detection systems. Supports three modes - basic (fast HTTP), stealth (undetectable), dynamic (browser automation).
0 -
knownasnaffy Bundle PDF Zusammenfuegen Workflow 3Datenschutzorientierter Workflow zum pdf zusammenfügen, pdf verbinden und Erstellen einer bewerbung pdf für deutschsprachige Nutzer. Verwende diesen Skill bei Anfragen zu PDF-Reihenfolge, Zusammenführung, Bewerbungsunterlagen, DSGVO-Anforderungen und sicheren Browser-Tools.
0 -
knownasnaffy Bundle Session History 2Search and browse past conversation history across all sessions. Use when recalling prior work, finding old discussions, resuming dropped threads, or when the user references something from a previous conversation that isn't in memory files. Also use when asked to "remember" something discussed before, find "that conversation about X", or continue work from a past session.
0 -
knownasnaffy Bundle Cc Statusline 2Configure Claude Code statuslines / 状态栏. Use this skill whenever the user wants to install, set up, preview, switch, clean up, customize, restore, or fix a Claude Code statusline: choose Full / Standard / Minimal / Developer presets, fine-tune an existing preset, generate-only or activate a custom 1 / 2 / 3-line layout with modules like model / git / context / token / cost, change theme or icon style, restore the default statusline while keeping scripts, or troubleshoot a broken statusLine command on Windows, macOS, or Linux. 也适用于安装、预览、切换预设、微调、美化、恢复默认、排障、自定义。 Only for Claude Code, not shell prompts, VS Code, Windows taskbar, web status bars, or Claude Desktop.
0 -
knownasnaffy Bundle Self Improvement 6Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.
0 -
knownasnaffy Bundle Double Agent 7This skill should be used when designing, implementing, or improving any AI system that requires quality assurance through separation of generation and evaluation roles. It implements the Generator-Evaluator dual-agent architecture (inspired by Anthropic's engineering blog and GAN design), where a Generator produces outputs and a dedicated Evaluator agent independently validates them through real interaction (e.g., Playwright browser operations), eliminating AI self-evaluation bias. Use when: building AI-generated UIs/code that needs quality checks, designing multi-agent pipelines with feedback loops, or upgrading existing coder+tester workflows to real interaction-based evaluation.
0 -
knownasnaffy Bundle China Consumer Electronics Suppliers 2Comprehensive consumer electronics industry suppliers guide for international buyers – provides detailed information about China's smartphone, wearable, audio, smart home, and AR/VR manufacturing clusters, supply chain structure, regional specializations, and industry trends (2026 updated).
0 -
knownasnaffy Bundle Feishu Group Thread Reply 5Force openclaw-lark bot replies into message threads in Feishu group chats, preventing main chat noise. Patches the plugin's dispatch layer and optionally the feishu-live-card watcher. Use when: (1) Bot replies appear in the main group chat stream instead of threads (2) After updating openclaw-lark plugin (patches get overwritten by npm updates) (3) User mentions "thread reply", "群聊 thread", "回复到话题", "thread 回复" (4) Checking if the thread reply patch is still applied (5) Setting up a new OpenClaw instance with Feishu group chats
0 -
knownasnaffy Bundle Byted Seedream Image Generate 4Generate high-quality images from text prompts using Volcano Engine Seedream models. Supports multiple artistic styles and aspect ratios. Use this skill when users want to create images from text descriptions, generate artwork in various styles, create visual content for creative projects, or need AI-powered image generation capabilities.
0 -
knownasnaffy Bundle Pyautogui 5Desktop automation via PyAutoGUI. Use when: user needs to automate mouse/keyboard actions, GUI testing, click/type sequences, screen-based workflows, or repetitive desktop tasks. NOT for: web automation (use Playwright/Selenium), mobile automation, or image recognition at scale.
0 -
knownasnaffy Bundle Psyche Engine 5Emergent psychological affect system — processes events into mood, trauma, attachment, romantic, humor, social awareness, role dynamics, consciousness index, hurt/recovery, and personality. Outputs a compact affect snapshot for behavior modulation.
0 -
knownasnaffy Bundle Polymarket 11Query Polymarket prediction market odds and events via CLI. Search for markets, get current prices, list events by category. Supports sports betting (NFL, NBA, soccer/EPL, Champions League), politics, crypto, elections, geopolitics. Real money markets = more accurate than polls. No API key required. Use when asked about odds, probabilities, predictions, or "what are the chances of X".
0 -
knownasnaffy Bundle Ppt Light Corporate 3Generate polished corporate PPTs using a light-themed company template. Activate when: (1) User asks to create/make/generate a PPT/PowerPoint/幻灯片/演示文稿 (2) User provides content and wants it turned into slides (3) User asks to improve or redo a PPT Uses a pre-analyzed H3C-style light corporate template (white bg + China red accent). NOT for: dark theme PPTs, creative/artistic decks, or Keynote files.
0
Frequently asked questions
What are AI & ML agent skills?
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
Which AI & ML skills are most installed?
Popular AI & ML skills on SkillMD right now include save-token, Polymarket, polymarket-candle-marubozu-trader. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do AI & ML skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.