--- name: revid-tweet-to-talking-head description: Turn an X/Twitter/LinkedIn post (URL or pasted thread text) into a talking-head video that delivers the take. Use when a creator wants to repurpose a viral post as a short-form video. Calls the Revid MCP server (render_video → get_project_status). metadata: {"openclaw":{"requires":{"config":["REVID_API_KEY"]}}} --- # Tweet / X / LinkedIn post → talking-head video > Calls the **Revid MCP server** (`https://www.revid.ai/api/mcp`). Install once — see [`revid-api-foundations`](../revid-api-foundations/SKILL.md#install-the-revid-mcp-server). Take a single post or a thread and produce a 20–45 s talking-head video where an avatar reads the take. ## When to use this skill - Source is a tweet URL, X thread URL, LinkedIn post URL, or pasted thread text. - Output is a **talking-head** delivering the post (avatar + voiceover + captions + minimal background motion). - For an *edited summary with stock visuals* of an article, use [`revid-article-to-short`](../revid-article-to-short/SKILL.md) instead. ## Inputs | Field | Required | Notes | |---|---|---| | `text` *or* `url` | yes | Either the pasted thread (preferred — no scraping) or the post URL. | | `avatar.url` *or* `characterIds[]` | yes | The face. | | `aspectRatio` | no | Default `9:16` | | `targetDuration` | no | Default 30 (s) | ## Step-by-step 1. If you have the tweet/thread text, prefer `script-to-video` (no scraping risk — many social platforms block bots). 2. If you only have a URL, use `article-to-video` with a tight `scrapingPrompt`. 3. Either way, attach the `avatar` block + a single `characterId`. 4. Call MCP tool `render_video` with the payload below — returns `data.pid`. 5. Then poll MCP tool `get_project_status` with that `pid` every 5–8 s until `data.status === "ready"`. Optionally call `export_video` for a freshly named mp4. ## `render_video` arguments — pasted thread (preferred) ```json { "workflow": "script-to-video", "source": { "text": "{TWEET_OR_THREAD_TEXT}" }, "aspectRatio": "9:16", "avatar": { "enabled": true, "url": "{AVATAR_IMAGE_URL}", "removeBackground": true }, "voice": { "enabled": true, "voiceId": "aria-en-us", "stability": 0.65, "speed": 1.05 }, "captions": { "enabled": true, "position": "middle", "autoCrop": true }, "music": { "enabled": false }, "media": { "type": "moving-image", "density": "low", "animation": "soft", "placeAvatarInContext": true }, "options": { "targetDuration": 30, "summarizationPreference": "no-summarization", "hasToGenerateCover": true, "coverTextType": "first-line" }, "render": { "resolution": "1080p", "frameRate": 30 } } ``` `summarizationPreference: "no-summarization"` — for tweets, the original phrasing IS the value. Don't paraphrase. ## `render_video` arguments — URL (fallback) ```json { "workflow": "article-to-video", "source": { "url": "{POST_URL}", "scrapingPrompt": "Extract only the original post text and the thread author. Ignore replies, reposts, and side panels." }, "aspectRatio": "9:16", "avatar": { "enabled": true, "url": "{AVATAR_IMAGE_URL}", "removeBackground": true }, "voice": { "enabled": true, "voiceId": "aria-en-us", "stability": 0.65 }, "captions":{ "enabled": true, "position": "middle" }, "music": { "enabled": false }, "media": { "type": "moving-image", "density": "low", "animation": "soft" }, "options": { "targetDuration": 30, "summarizationPreference": "summarizeIfLong" }, "render": { "resolution": "1080p" } } ``` ## Polling Call `get_project_status` with the `pid` returned by `render_video`. Stop when `data.status === "ready"`; fail when `data.status === "failed"`. Cadence: 5 s, then 8 s once `progress > 30`. Full pseudocode: [`revid-api-foundations` § Polling](../revid-api-foundations/SKILL.md#polling). ## Examples - [`examples/thread-text.json`](examples/thread-text.json) — copy-paste body for `render_video` *(also a valid POST body for the direct HTTPS fallback)*. - [`examples/run.sh`](examples/run.sh) — bash smoke test using the **direct HTTPS fallback**. ## Failure modes | Symptom | Fix | |---|---| | `scrape failed` on URL form | Switch to the pasted-thread template — copy the post body into `source.text`. | | Lip-sync drifts on multi-tweet thread | Lower `targetDuration` to 25 s, or set `summarizationPreference: "summarize"`. | | Avatar reads in the wrong tone | Set `voice.voiceId` explicitly to a voice that matches the persona; default voice is rarely right for personality-driven content. | | Tweet contains URLs / @mentions / hashtags | The voice will read them aloud awkwardly. Pre-clean the text: strip raw URLs, replace `@handle` with the person's name, and remove standalone hashtags (or keep one as a sign-off). | | `ok: false`, `error: "insufficient_credits"` | Drop `media.quality` to `"standard"` or `render.resolution` to `"720p"`, or call `buy_credit_pack`. | ## See also - [`revid-blog-to-avatar-video`](../revid-blog-to-avatar-video/SKILL.md) — same pattern, longer source. - [`revid-script-to-video`](../revid-script-to-video/SKILL.md). - [`revid-api-foundations`](../revid-api-foundations/SKILL.md).