Menu
Animate a still image into a short video via OpenRouter. Default model google/veo-3.1-fast.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | openrouter-image2video |
| description | Animate a still image into a short video via OpenRouter. Default model google/veo-3.1-fast. |
Image + motion-prompt → video via OpenRouter. Default model: google/veo-3.1-fast.
Use this when you already have a strong still image and want to bring it to life with subtle motion (camera pan, parallax, gentle animation) while preserving the composition. For motion-from-scratch, use openrouter-text2video instead.
The script accepts either a remote image URL or a local image path; local files are base64-encoded and inlined into the request as a data URL.
Resolve TOOL_DIR = the directory containing this SKILL.md. Commands below use TOOL_DIR as a symbolic placeholder; replace it with the resolved, quoted path before running Bash.
export OPENROUTER_API_KEY=sk-or-v1-...
# From a local image you already generated with text2image
python3 TOOL_DIR/scripts/generate_video_from_image.py \
--image PROJECT_DIR/assets/teaser.png \
--prompt "slow parallax push-in, soft drift of ambient particles, no camera shake" \
--duration 5 \
--aspect-ratio 16:9 \
--download PROJECT_DIR/assets/teaser.mp4
# Or from a remote URL
python3 TOOL_DIR/scripts/generate_video_from_image.py \
--image-url "https://example.com/still.png" \
--prompt "subtle camera dolly forward, gentle depth-of-field shift" \
--download PROJECT_DIR/assets/scene.mp4
| Flag | Default | Description |
|---|---|---|
--prompt | required | Motion prompt — describe what should move and how |
--download | required | Output MP4 path |
--image | one of --image / --image-url required | Local image path (PNG/JPG); will be base64-encoded |
--image-url | one of --image / --image-url required | Remote image URL |
--model | google/veo-3.1-fast | Any OpenRouter image-to-video-capable model |
--duration | 5 | Seconds |
--aspect-ratio | 16:9 | 16:9, 9:16, 1:1, 4:3, 3:4, 21:9, 9:21 |
--resolution | 720p | Model-dependent (e.g. 480p, 720p, 1080p) |
--frame-role | first | first or last — anchor frame role for the input image |
--generate-audio | off | Generate audio with video (if model supports) |
--poll-interval | 5 | Seconds between polls |
--max-wait | 600 | Max total wait time |
POST /api/v1/videos with body:
{
"model": "google/veo-3.1-fast",
"prompt": "...motion prompt...",
"aspect_ratio": "16:9",
"duration": 5,
"resolution": "720p",
"frame_images": [
{
"type": "image_url",
"frame_type": "first_frame",
"image_url": {"url": "data:image/png;base64,..." }
}
]
}
--frame-role first|last flag maps to frame_type: "first_frame"|"last_frame".GET /api/v1/videos/{id} every 5s until status == "completed"GET /api/v1/videos/{id}/content → raw MP4 bytesframe_images with roles first and last. The current script wires only one anchor frame; extend body["frame_images"] to add a second.Exhaustively profile a dataset and list ALL possible analyses — distributions, correlations, rankings, trends, group comparisons, anomalies. Reads detective.json for context. Runs after the Detective/Scout, before the Editor. Outputs analyst.json with ana_xx IDs and chart-ready data_tables.
Audit a generated Data2Story blog for build correctness across ALL modalities by ACTUALLY RENDERING it in a real headless browser (when available) — catching blank/0-width charts, broken/oversized media, desktop+mobile overflow, and dead interactions that source inspection cannot see — then having vision agents review the screenshots, playtesting every interactive, and walking the flagship capability contract. Plus VIEW every image for a wrong/AI-faked subject. Fixes layout in place; sends content-correctness problems back to the owning role. Use at Stage 6 after the Programmer assembles index.html and before the Critic; re-runs each revision round. Triggers: a built index.html exists, or you need to confirm charts/media/interactions actually render.
Design the MANDATORY cinematic scroll experience (every blog ships one): a full-bleed, scroll-driven background — primarily from the Scout's verified real imagery — plus the global motion choreography, so the narrative unfolds like a film as the reader scrolls. Owns the BACKGROUND layer + page-level motion (not the Designer's per-section visuals, not the Interaction centerpiece — it stages them inside the scroll). Outputs cinematographer.json (cin_xx scenes). There is no off path: a topic with no sourceable real imagery falls back to a generative-atmosphere or data-driven-spine background, never to a bare column.
Name the piece — re-write the masthead (headline + standfirst + kicker), every section title, and every figure/photo/table caption to a research-driven titling standard, killing the AI-tell patterns (the 'flat statement. flat counter-statement.' two-beat above all) a competent default falls into. Reads editor.md/json + analyst.json + the resolved topic_profile; writes copywriter.json — STRINGS ONLY (masthead{headline,standfirst,kicker}, items{edt_xx:{title}, des_xx:{caption}}), each backed by a real ana_*. Names, never edits: it touches no finding, no number, no data-* id, no layout — so the Verify layer is untouched and the Programmer renders the masthead + figcaptions from copywriter.json verbatim. Runs at Stage 3.5, after the Editor, before the Designer.
Review a finished Data2Story blog against the 5 quality rubric dimensions (visual_design, narrative_pacing, data_method_transparency, claim_data_alignment, insight_value), score each 1-7 with on-page evidence, and emit critic.json with pass/fail + targeted, surgical send-back instructions. Verifies every load-bearing claim/asset against its traceability chain before scoring; applies the caveat-survival, honest-accuracy, and third-party-attribution caps. Does NOT rewrite content — scores and sends back. Use at Stage 6.5 after the Auditor and after verify.py has produced verifier.json; re-runs each revision round. Triggers: a built index.html plus verifier.json exist, or you need to judge whether the article is actually good.
Generate text embeddings via OpenRouter using Qwen3-Embedding-8B.