# Send TED Talk to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "ted-talk", "name": "TED Talk", "source": "tencent", "type": "skill", "category": "内容创作", "sourceUrl": "https://clawhub.ai/leegitw/ted-talk", "canonicalUrl": "https://clawhub.ai/leegitw/ted-talk", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/ted-talk", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=ted-talk", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "SKILL.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "ted-talk", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T18:35:17.839Z", "expiresAt": "2026-05-07T18:35:17.839Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=ted-talk", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=ted-talk", "contentDisposition": "attachment; filename=\"ted-talk-1.0.2.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "ted-talk" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/ted-talk" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/ted-talk", "downloadUrl": "https://openagent3.xyz/downloads/ted-talk", "agentUrl": "https://openagent3.xyz/skills/ted-talk/agent", "manifestUrl": "https://openagent3.xyz/skills/ted-talk/agent.json", "briefUrl": "https://openagent3.xyz/skills/ted-talk/agent.md" } } ``` ## Documentation ### ted-talk (話) Transform technical conversations and insights into full-length TED-style talks. Creates comprehensive 40-50 minute presentations with hooks, concrete examples, broader implications, and Q&A preparation. Trigger: 明示呼出 (explicit invocation) or when deep technical insight emerges Core insight: "The best technical talks don't just explain what — they reveal why it matters, with examples concrete enough to apply and implications broad enough to inspire." ### Installation openclaw install leegitw/ted-talk Dependencies: None (standalone creative skill) Data handling: This skill synthesizes content from user-supplied input or the current conversation context (default). It does NOT read files from the workspace or access project artifacts directly. Results are returned to the invoking agent, who decides how to use them. ### What This Solves Technical insights often stay trapped in conversations. A well-structured talk makes them teachable and shareable. This skill: Expands technical conversations into comprehensive narratives Grounds abstract insights in concrete, real-world examples Prepares for audience questions and objections The insight: A 50-minute talk forces you to truly understand something — if you can't explain the why, address objections, and connect to broader implications, you don't fully understand it. ### Usage /ted [topic] ### Arguments ArgumentRequiredDescriptiontopicNoTopic focus (default: synthesize current conversation) ### Pre-Requisites Before creating a TED talk, ensure: Sufficient conversation depth — Surface-level topics make shallow talks Clear narrative arc — Problem → Discovery → Solution → Impact Main work documented — Save current progress first if mid-task Concrete context available — Real problems, real decisions, real outcomes ### Context Understanding Checklist You must be able to answer: QuestionWhat It MeansCore insight?Not "we talked about X" but "we discovered X solves Y"Problem solved?The pain point, not just the topicWhy, not just what?The reasoning, not just the outcomeConcrete examples?Specific details from the context providedBroader implications?Why does this matter beyond the immediate context? ### Red Flags (Don't Create) Surface-level summary of conversation Don't understand why a decision was made No concrete examples to draw from Insight doesn't have broader implications Would be padding the talk with generic content ### TED Talk Structure ## TED Talk: "[Talk Title]" ### Opening (0:00-2:00) [Hook with relatable problem] ### Setup: Why This Matters (2:00-6:00) [Context, stakes, personal connection] ### The Problem (6:00-12:00) [Deep dive into the pain point] ### Core Concept (12:00-25:00) [Explain the insight thoroughly] ### Real-World Examples (25:00-38:00) [Concrete applications with specifics] ### Broader Implications (38:00-45:00) [Why this matters beyond immediate context] ### Closing (45:00-48:00) [Call to action or reflection] ### Q&A Preparation (48:00-50:00) [Common objections and responses] ### Talk Rules RuleDescriptionFull 40-50 minutesNOT a summary — comprehensive contentHook with problemStart with relatable pain, not abstract conceptConcrete examplesSpecific, real details — not hypothetical scenariosAddress objectionsQ&A section anticipates pushbackNo fillerEvery section should teach somethingAccessible but not condescendingTechnical depth without jargon overload ### Example: Section Length Guide SectionDurationPurposeOpening2 minHook with relatable problemSetup4 minWhy this matters, stakesProblem6 minDeep dive into pain pointCore Concept13 minThe main insight, thoroughly explainedExamples13 minReal-world applicationsImplications7 minBroader impactClosing3 minCall to actionQ&A Prep2 minObjections and responses ### Step 1: Synthesize Conversation Read full conversation context Identify key decisions, "aha" moments Extract core insight or pattern Note concrete details from the context provided ### Step 2: Identify Narrative Arc ElementQuestionProblemWhat was broken/painful?DiscoveryWhat did we learn?SolutionWhat pattern emerged?ImpactWhy does this matter? ### Step 3: Expand to Talk For each section: Opening: What relatable problem hooks the audience? Setup: Why should they care? What's at stake? Problem: Deep dive — make them feel the pain Concept: Explain thoroughly, with analogies if helpful Examples: Specific, concrete, from real work Implications: Connect to broader context Closing: What should they do with this knowledge? Q&A: What will skeptics ask? ### Step 4: Ground in Reality TED talks resonate when they're specific, not hypothetical. Draw from: Real problems encountered (not abstract scenarios) Actual decisions and their reasoning Specific outcomes and what changed Stories with concrete details the audience can visualize Use what the user provides — don't invent specifics or assume access to files. ### Input: Bootstrap Observability Insight Context: Discovered reproduce-to-debug problem. Introduced Bootstrap → Learn → Enforce phases. ### Output Excerpt Title: "Bootstrap Before You Break: Why Greenfield Systems Need to Learn What Normal Looks Like" ### Opening (0:00-2:00) It's 3 AM. Your pager goes off. The system is on fire — users are complaining, something is clearly wrong. You pull up your logs and... nothing. Metrics? Flat lines. Traces? What traces? You're debugging blind. And the worst part? You built this system. You know every line of code. But you have no idea what's happening. This happened to me six months ago. And the solution we found changed how I think about every system I'll ever build. ### Setup: Why This Matters (2:00-6:00) We've all been taught: instrument your code, add logging, set up dashboards. The standard advice works great — for existing systems. But what about greenfield? What about day one? Here's the thing nobody tells you: you can't set meaningful thresholds for a system that's never run in production. What's a normal response time? What's an acceptable error rate? You don't know. You can't know. The system hasn't told you yet. [... continues for full 50 minutes ...] Note: Full TED talk outputs are 40-50 minutes of content. This excerpt demonstrates the opening sections. ### Q&A Preparation Q: "This seems like overkill for small projects." A: Fair point. The full three-phase approach is designed for systems where debugging cost is high — production services, distributed systems, anything where "just add a log line and redeploy" isn't an option. For a weekend project, you probably don't need this. But the core insight — bootstrap before you enforce — scales down too. Even a simple log.Debug everywhere is a form of bootstrap phase. Q: "How long should the bootstrap phase last?" A: We found 7-14 days covers most patterns. You want at least one full business cycle (weekly patterns), and ideally two. The key indicator is when your anomaly detection stops alerting on normal behavior. ### Integration Layer: Creative Depends on: None (standalone) Used by: side-quests (combo skill) Complements: insight-song, visual-concept ### Failure Modes ConditionBehaviorInsufficient contextAsk clarifying questions firstNo concrete contextAsk for specific details before proceedingSurface-level insightSuggest deeper exploration firstNo broader implicationsSuggest finding wider relevance ### Security Considerations Input sources: User-supplied context (if provided) Current conversation context (default) What this skill does NOT do: Read files from the workspace Access project artifacts directly Send data to external services Record or publish content Output behavior: This skill returns the full TED talk directly to the invoking agent. The agent can then display, save, or pass the result to another skill as needed. Note on concrete details: The skill uses only what the user provides in the conversation or as explicit input. It does not access workspace files. Review output before sharing externally to ensure no sensitive information is included. Provenance note: This skill is developed by Live Neon (https://github.com/live-neon/skills) and published to ClawHub under the leegitw account. Both refer to the same maintainer. ### Quality Checklist Can explain core insight in one sentence Opening hooks with relatable problem Full 40-50 minutes of substantial content Concrete details from provided context included Broader implications explored Q&A addresses likely objections No filler or generic padding Accessible to non-experts but not condescending ### Acceptance Criteria /ted synthesizes input or conversation into full-length talk Output includes all sections (opening through Q&A) Talk is 40-50 minutes of substantial content Concrete details from input/conversation included Q&A section addresses common objections Result returned to invoking agent Part of the Live Neon Creative Suite. ## Trust - Source: tencent - Verification: Indexed source record - Publisher: leegitw - Version: 1.0.2 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-04-30T18:35:17.839Z - Expires at: 2026-05-07T18:35:17.839Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/ted-talk) - [Send to Agent page](https://openagent3.xyz/skills/ted-talk/agent) - [JSON manifest](https://openagent3.xyz/skills/ted-talk/agent.json) - [Markdown brief](https://openagent3.xyz/skills/ted-talk/agent.md) - [Download page](https://openagent3.xyz/downloads/ted-talk)