Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add 62656456/ai-film-skills --skill produce-ai-videogit clone --depth 1 https://github.com/62656456/ai-film-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/62656456/ai-film-skills/produce-ai-video)<a href="https://agentmods.dev/skills/62656456/ai-film-skills/produce-ai-video"><img src="https://agentmods.dev/badge/skills/62656456/ai-film-skills/produce-ai-video/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/62656456/ai-film-skills/produce-ai-video"><img src="https://agentmods.dev/badge/skills/62656456/ai-film-skills/produce-ai-video.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00110 | $0.01301 |
| Opus 5 | $0.00055 | $0.00651 |
| Sonnet 5 | $0.00022 | $0.00260 |
| Haiku 4.5 | $0.00011 | $0.00130 |
Grade A, and why
produce-ai-video scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Produce AI Video
Outcome
Deliver an actual qualified video. Treat storyboards, prompts, generated clips, edit projects, and QC reports as intermediate artifacts, never as the final result.
This package contains its own production storyboard and six-module prompt compiler in references/storyboard-prompt-compiler.md. Use it only after the director judgment is complete.
Choose the execution mode
Autonomous production mode
Use this mode when the user asks Codex to create autonomously, automatically make the video, test its ability, or deliver a finished film from a script.
- Read the project knowledge, current script, approved assets, visual rules, decisions, and checkpoints.
- Make the director decisions independently.
- Design the shot groups, prompts, production route, generation, selection, edit, sound, review, and repairs.
- Do not push ordinary directing decisions back to the user.
- Ask only when a missing choice materially changes story meaning, spend, permissions, publication, or an already locked structure.
User-directed execution mode
Use this mode when the user supplies or has repeatedly adjusted a storyboard, timing, shot order, staging, dialogue, or prompt structure.
- Treat the user-locked structure as authoritative.
- Change only the requested scope.
- Do not add shots, remove shots, reorder beats, rewrite dialogue, or replace staging in the name of optimization.
- If a requested result conflicts with the locked structure, identify the exact conflict and its visible consequence before proposing a change.
Run the autonomous workflow
- Lock sources and acceptance. Identify the unique project, current script version, approved assets, fixed decisions, delivery format, permissions, cost boundary, and definition of a qualified video. Separate verified facts, unknowns, assumptions, and conventions.
- Interpret the script. Determine the dramatic event, character objective, power relation, information reveal, physical action, emotional turn, sound cue, entry state, and exit state. Read autonomous-production-workflow.md for the auditable decision framework.
- Direct before prompting. Decide what the audience must see and in what order. Build a world-state model for space, subjects, props, light sources, movement axes, and continuity.
- Design segments and shots. Treat a segment as a short dramatic sequence and a shot as one uninterrupted viewpoint. In autonomous mode, design at least 7–8 effective, non-equally timed shots per segment unless the user explicitly requests a long take. Every cut must add information, change power, clarify action, reveal a reaction, or hand off the next beat.
- Create and compile the director package. Produce a
DIRECTOR_SHOT_PACKAGEcontaining the segment objective, entry and exit states, world-state lock, shot order, timing, framing, camera, visible action, sound, cut motivation, and continuity handoff. Only after this package is coherent mayreferences/storyboard-prompt-compiler.mdconvert it into the approved five-column storyboard and six-module video prompt. - Choose the production route. Preserve the director timing and shot design. If one model call cannot reliably render the required internal shots, generate individual shots or smaller clusters and edit them into the designed segment. Never let a model's maximum duration redefine the dramatic timing.
- Generate real motion. Produce actual video material. Reject static-frame motion, keyframe slideshows, or technical previews when the requested deliverable is a finished video.
- Select and assemble. Judge takes by performance, identity, action, continuity, composition, and editability. Cut on motivated action, gaze, occlusion, object, sound, or information change. Add handles where the tool permits; do not concatenate fixed clip durations blindly.
- Build sound. Integrate dialogue, performance breaths, environment, effects, transitions, silence, and music only when authorized. Make sound carry space, action, rhythm, and continuity rather than feeling pasted on.
- Watch, repair, and rewatch. Review the entire film at normal speed for story and rhythm, then again for continuity, artifacts, and sound. Fix the first audience-rejecting defect and repeat until all hard gates in qualified-video-acceptance.md pass.
- Deliver honestly. Return the playable final video, its duration and format, the validation state, and any visible residual risk. If full-playback review or a hard gate is unavailable, report
未完成/待验证; never call the result qualified.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago Changed 05ce3b602907
- 12d ago First seen · 68 lines · 110 tokens per session scan A e2721e4d6493
produce-ai-video is a skill published in the GitHub repository 62656456/ai-film-skills (17 stars, last pushed yesterday), licensed Apache-2.0. It adds 110 tokens to every session and 1,301 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
seedance-20
This skill should be used when creating, improving, or troubleshooting Seedance 2.0 video on any surface — Dreamina, Jimeng, CapCut, Doubao, Volcengine/Ark, BytePlus, Runway's Seedance route, fal, or third-party provider/router surfaces such as EvoLink, OpenRouter, Kie.ai, PiAPI, LaoZhang, Runware, ModelsLab, AI/ML…
seedance-prompt
This skill should be used when the user asks to write, improve, translate, compress, or debug a Seedance 2.0 video prompt; mentions T2V, I2V, V2V, R2V, camera direction, prompt quality, or provides reference assets for a production-ready prompt.
script-to-shootable-storyboard
A guide for turning scripts, story outlines, dialogue, or existing storyboards into detailed shot plans. A storyboard is a sequence of planned camera shots showing what viewers see and hear.
travel-skill
A production guide for realistic travel-promotion videos that combine authorised real footage with AI-generated shots.
minimax-h3
Write, debug and structure prompts for MiniMax H3 video generation (T2VA, I2VA, FL2VA, L2VA, Ref2VA) and configure its ComfyUI workflow. Use when the user mentions MiniMax H3, minimaxh3, fl2va, ref2va, MiniMaxH3ReferenceToVideo, reference-to-video, asks to animate a photo, write a video prompt, pick a model quant, or…
minimax-h3-prompting
Guide an idea into a generation-ready MiniMax H3 / Hailuo H3 prompt or diagnose an inspected H3 output. Use for T2VA, I2VA, FL2VA, L2VA, or Ref2VA; text-only PV and kinetic type, Motion Design/MG, packaging and transitions, product/UI/game/MV/title work, localized reality-to-hand-drawn edits, audio/timbre reference…