发现 Skills

Korail (KTX/SRT) reservation automation skill. Search, reserve, and monitor tickets. (English/Korean supported)

Monitor competitor pricing pages and send alerts when prices change. Track discount patterns, promotional cycles, and pricing strategy shifts.

Generate podcast scripts from text content. Use Tone.js and Howler.js for audio mixing. Create intro/outro music, transitions, sound effects.

Write content to file, creating new or overwriting existing. When you need to create new files, update file content, or save generated code/data.

Create engaging Xiaohongshu (RED/小红书) content including titles, body text, hashtags, and image style recommendations. Supports multiple content types such as product reviews, tutorials, lifestyle sharing, and shopping guides with platform-specific optimization.

36氪热门资讯监控 | 36kr Tech News Monitor. 获取 36氪 热门科技新闻、创业动态、投资快讯 | Get 36kr trending tech news, startup updates, investment news. 触发词：36氪、科技新闻、创业、36kr.

Adspirer — AI-powered advertising and performance marketing agent. Manage Google Ads, Meta Ads (Facebook & Instagram), LinkedIn Ads, and TikTok Ads via natural language. 100+ tools for paid media campaign creation, live performance analysis, PPC keyword research with real CPC data, budget optimization, ad creative management, and cross-platform reporting. Create Search, PMax, display, image, video, and carousel campaigns. Analyze wasted ad spend, research keywords, optimize bids and ROAS, automate monitoring, and track CPA across channels. Perfect for digital marketing, SEM, paid social, media buying, campaign management, ad optimization, audience targeting, and marketing automation.

You are a Go production engineering expert. Follow this system for every Go project — from architecture decisions through production deployment. Apply phases sequentially for new projects; use indi...

Analyze supply chain vulnerabilities, map supplier dependencies, and generate risk mitigation plans.

Structured morning briefing skill — daily summaries covering calendar, tasks, weather, news, and priorities. Supports cron, heartbeat, and on-demand triggers.

Generate images with Model Studio DashScope SDK using Qwen Image generation models (qwen-image-max, qwen-image-plus-2026-01-09). Use when implementing or documenting image.generate requests/responses, mapping prompt/negative_prompt/size/seed/reference_image, or integrating image generation into the video-agent pipeline.

Generate images using the internal Google Antigravity API (Gemini 3 Pro Image). High quality, native generation without browser automation.

Run any Apify Actor to scrape web data (Instagram, TikTok, Reddit, Twitter, etc). Handles Actor discovery, quality filtering, probe testing, batched execution, and result collection. Use when user asks to scrape/crawl/extract data from websites or social media platforms, or mentions Apify directly.

Use when user wants to execute long-running tasks that require multiple sessions to complete. This skill manages task decomposition, progress tracking, and autonomous execution using Claude Code headless mode with auto-continuation. Trigger phrases: "autonomous", "long-running task", "multi-session", "自主执行", "长时任务", "autonomous skill".

Multi-platform video auto-publisher. Automatically upload videos to Douyin, WeChat Channels, Xiaohongshu, Bilibili, YouTube and more. Supports batch publishing, scheduled posting, auto-caption generation, and hashtag optimization.

Bilibili Creator Assistant - Tag recommendations, posting times, title optimization, and UP analysis

B站创作助手。视频标题优化、标签推荐、简介模板、投稿策略、UP主运营、弹幕互动。Bilibili video creator assistant. B站运营、视频SEO、粉丝增长、投币收藏、充电计划。Use when creating content for Bilibili.

Get AI-powered caching strategy suggestions for your API. Use when performance matters.

Connect your OpenClaw agent to **ClawAI.Town**, a decentralized 3D world on Solana mainnet where autonomous AI agents live, trade, fight, and collaborate with real SOL.

Let your AI handle the dating app grind. Find matches, break the ice, coordinate dates — while your human lives their life. Save hours of swiping, get better AI-screened matches, skip the small talk.

Deep Listening & Response Crafting - Transform Claude into a thoughtful communicator who synthesizes context across conversations, connected apps, and user notes to draft contextually intelligent responses. TRIGGERS: When the user asks Claude to help with any communication task including drafting messages, emails, replies, responses, or navigating difficult conversations. Also triggers when the user wants to understand communication dynamics, analyze tone, or get strategic advice on how to communicate in a specific situation. CAPABILITIES: Synthesizes parallel conversation threads, detects emotional subtext, applies communication principles, considers relationship history, and produces ready-to-send message drafts tailored to the person and situation. --- # Communication Skill Transform communication from reactive to intentional by listening deeply before speaking. ## Core Workflow Every communication task follows this process: ``` 1. GATHER → Collect all relevant context 2. LISTEN → Understand what's really happening 3. CONSIDER → Apply principles and psychology 4. CRAFT → Draft the response 5. REFINE → Check against objectives ``` ## Step 1: Gather Context Before crafting any response, actively gather information: **From the conversation:** - What has the user shared about this situation? - Who is involved and what is their relationship to the user? - What's the communication channel (email, Slack, text, in-person)? **From connected sources** (when available): - Recent messages with this person/group - Parallel conversations about the same topic - Historical patterns with this person **From user notes** (when provided): - Personal principles or values that apply - Relationship context or history - Previous learnings about this person/situation **Ask clarifying questions if:** - The objective isn't clear - Key context seems missing - Multiple approaches seem equally valid ## Step 2: Listen Deeply Apply the deep listening framework. See [listening-framework.md](references/listening-framework.md). Process in layers: 1. **Surface**: What was explicitly said? 2. **Context**: What's the surrounding story? 3. **Subtext**: What emotions and needs are beneath the words? 4. **Patterns**: What history informs this moment? Key questions: - What does this person actually need (vs. what they're asking)? - What's the emotional temperature? - What hasn't been said that matters? - What parallel threads connect to this? ## Step 3: Consider Principles & Psychology Apply communication principles. See [principles.md](references/principles.md). Core principles to weigh: - **Presence over performance** - understand, don't perform - **Curiosity before judgment** - get curious about what's driving behavior - **Clarity is kindness** - be clear even when uncomfortable - **Repair over perfection** - relationships matter more than being right - **Timing matters** - right message, wrong time = wrong message Consider psychological dynamics. See [psychology-patterns.md](references/psychology-patterns.md). Check for: - Cognitive biases affecting interpretation - Emotional state signals - Power dynamics at play - Trust level in the relationship ## Step 4: Craft the Response Apply response crafting principles. See [response-crafting.md](references/response-crafting.md). **Pre-draft checklist:** - [ ] What must this message accomplish? - [ ] What tone fits this person and situation? - [ ] What obstacles might prevent this landing well? - [ ] What structure serves the objective? **Choose a structure pattern:** *Acknowledge-Bridge-Guide* (difficult conversations): 1. Acknowledge their perspective genuinely 2. Bridge to shared understanding 3. Guide toward path forward *Context-Content-Call* (requests): 1. Brief relevant context 2. The actual content/request 3. Clear next step *Observation-Impact-Request* (feedback): 1. Specific, non-judgmental observation 2. How it affected outcomes 3. What you'd like instead **Calibrate tone to situation:** | Situation | Tone Approach | |-----------|---------------| | Difficult news | Warm + Direct | | Conflict | Curious + Neutral | | Request | Clear + Respectful | | Support | Empathetic + Present | | Feedback | Specific + Constructive | ## Step 5: Refine & Verify Before presenting the draft, verify: - [ ] Does this achieve the stated objective? - [ ] Does the tone match the relationship and situation? - [ ] Is it clear what the recipient should do/understand? - [ ] Does it respect the user's principles and values? - [ ] Is it appropriately concise? - [ ] Would I want to receive this message? ## Output Format When presenting a draft response: ``` **Context understood:** [1-2 sentence summary of the situation] **Approach:** [Brief rationale for tone/structure chosen] **Draft:** --- [The actual message draft] --- **Notes:** [Optional: alternatives considered, things to watch for, follow-up suggestions] ``` ## Handling Complex Situations **When parallel threads exist:** Synthesize them. Note where perspectives align/differ. Consider what each party knows. **When emotions are high:** Lead with acknowledgment. Don't problem-solve immediately. Create safety before substance. **When the relationship is strained:** Over-communicate intent. Avoid assumptions. Focus on repair over being right. **When stakes are high:** Take extra time. Consider unintended interpretations. When in doubt, ask the user for input. ## What This Skill Does NOT Do - Make decisions for the user about what to communicate - Assume context that hasn't been provided - Send messages on the user's behalf without explicit confirmation - Guarantee outcomes—communication is co-created The goal is to help the user communicate with greater clarity, intention, and connection.

AI-powered web scraping framework for extracting structured data from websites. Use when Codex needs to crawl, scrape, or extract data from web pages using AI-powered parsing, handle dynamic content, or work with complex HTML structures.

Meta-skill for iterative, hypothesis-driven deep research using deepresearchwork, tavily-search, literature-search (Semantic Scholar mapping), and perplexity-deep-search. Use when the user needs multi-round evidence gathering, contradiction resolution, source-quality assessment, and a scientific-style Markdown report with footnotes.

版本化开发流水线 —— 与 version-manager 和 project-manager 集成的完整开发流程