Plan draft video-ad variations from supplied performance evidence, preserving the distinction between observed data and creative hypotheses.
The English summary is being prepared. Showing the original description.
Establish durable project context, asset roles, delivery constraints, and approval state for repeatable video work.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Create the small amount of project context that prevents repeated video work from losing the brief, source roles, continuity rules, rights state, or delivery constraints. This is a planning and workspace skill; it does not submit paid media jobs.
Ask for or extract:
Do not convert an unverified statement into a fact. Keep unknown,
user-supplied, provider-observed, and approved distinct.
For each asset, record a stable local ID, type, role, source, parent if any,
rights/consent state, intended use, and whether it may be uploaded. Useful
roles include subject_reference, product_reference, first_frame,
last_frame, driving_video, source_video, audio, logo, script, and
approved_parent.
Do not infer that a file is safe to upload because it is present locally. Ask for permission when an asset contains a real person, private information, licensed material, or an externally owned brand.
When workspace writes are allowed, create or update .video/project.md using
the structure in reports and receipts.
Preserve prior approved facts and append a dated change note when a later
brief supersedes them. Do not overwrite a receipt or approved parent merely to
make the folder look tidy.
At minimum, the context should include:
If a project context already exists, summarize the differences and ask before changing an approved invariant. If writing is not authorized, return the proposed context in the response without claiming it was saved.
Return the project context path or draft, asset-role table, unresolved decisions, rights/consent gaps, and the recommended next skill. Typical next steps are video-generation, storyboard-to-video, character-continuity, or short-form-repurposing.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Plan draft video-ad variations from supplied performance evidence, preserving the distinction between observed data and creative hypotheses.
The English summary is being prepared. Showing the original description.
Create authorized presenter, avatar, and UGC-style video drafts with explicit voice, likeness, consent, and synthetic-media handling.
The English summary is being prepared. Showing the original description.
Analyze a script or existing edit, identify visual gaps, and generate supplemental B-roll coverage with clear parent and factual-status labels.
The English summary is being prepared. Showing the original description.
Build a consented continuity system for an authorized person, fictional character, product, or world across video shots.
The English summary is being prepared. Showing the original description.
Translate creative intent into coherent cinematography, framing, camera, lens, lighting, motion, and sound direction for a video shot.
The English summary is being prepared. Showing the original description.
Create a non-explicit romantic scene from authorized adult references with clear consent, identity boundaries, and synthetic-media labeling.
The English summary is being prepared. Showing the original description.