--- name: revid-pdf-to-video description: Turn a PDF (whitepaper, ebook chapter, slide deck export, research paper) into a short summary video. Use when the source is a PDF URL or a PDF the agent can upload to public storage first. Calls the Revid MCP server (render_video → get_project_status). metadata: {"openclaw":{"requires":{"config":["REVID_API_KEY"]}}} --- # PDF → summary 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 PDF URL and produce a short summary video. Internally this routes through `article-to-video` once the PDF text has been extracted by Revid's scraper. ## When to use this skill - Source is a public PDF URL (whitepaper, paper, ebook, slide export). - Goal is a 30–90 s summary, not a full reading. - For HTML articles use [`revid-article-to-short`](../revid-article-to-short/SKILL.md). - For local PDFs the agent must first upload to public storage (S3, Supabase Storage, etc.) so Revid can fetch it. ## Inputs | Field | Required | Notes | |---|---|---| | `url` | yes | Public PDF URL (must be reachable, no auth) | | `aspectRatio` | no | Default `9:16` | | `targetDuration` | no | Default 60 (s); raise to 90 s for dense papers | ## Step-by-step 1. Confirm the URL ends in `.pdf` or returns `Content-Type: application/pdf`. 2. Set `source.scrapingPrompt` to bias the summary toward what the user cares about ("Focus on the methodology section", "Focus on the executive summary", "Pull the 3 biggest takeaways"). 3. Call MCP tool `render_video` with the payload below — returns `data.pid`. 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 ```json { "workflow": "article-to-video", "source": { "url": "{PDF_URL}", "scrapingPrompt": "Extract the executive summary and the 3 biggest takeaways. Skip references and appendices." }, "aspectRatio": "9:16", "voice": { "enabled": true, "stability": 0.65, "speed": 0.95, "language": "en-US" }, "captions": { "enabled": true, "position": "middle", "autoCrop": true }, "music": { "enabled": true, "syncWith": "beats" }, "media": { "type": "stock-video", "density": "medium", "animation": "soft", "quality": "pro", "imageModel": "good", "videoModel": "pro" }, "options": { "targetDuration": 60, "summarizationPreference": "summarize", "soundEffects": true, "hasToGenerateCover": true, "coverTextType": "title" }, "render": { "resolution": "1080p", "frameRate": 30 } } ``` ## 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/whitepaper.json`](examples/whitepaper.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** (`POST /api/public/v3/render` → `GET /status`). Useful when you don't have an MCP client at hand. ## Failure modes | Symptom | Fix | |---|---| | `scrape failed` | PDF behind auth or login wall. Re-host the PDF on public storage. | | Summary skips the section the user cares about | Tighten `scrapingPrompt` (be specific: section names, page ranges). | | Visuals are abstract / off-topic | PDFs rarely have crawlable hero images. Pre-render a few key figures as images and pass them in `media.provided`. | | Voice rushes through technical terms | Lower `voice.speed` to `0.9`. | | `ok: false`, `error: "insufficient_credits"` | Drop `media.quality` to `"standard"` or `render.resolution` to `"720p"`, or call `buy_credit_pack`. | ## See also - [`revid-article-to-short`](../revid-article-to-short/SKILL.md) for HTML. - [`revid-script-with-custom-media`](../revid-script-with-custom-media/SKILL.md) for full visual control.