svenkatreddy/universal-video-skills / skills/model-adapters
model-adapters
VerifiedTranslate a finished shot-brief into per-model instructions for Muse, Gemini/Veo, Sora, Kling, Runway, Pika, Luma, and Hailuo. Constraint sheets, prompt quirks, and what to verify. Use after the prompt is written, before generating.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.2 KB |
| references/adapter-template.md | markdown | 962 B |
| references/gemini-veo.md | markdown | 1.7 KB |
| references/hailuo.md | markdown | 1.4 KB |
| references/kling.md | markdown | 1.2 KB |
| references/luma.md | markdown | 1.1 KB |
| references/muse.md | markdown | 1.6 KB |
| references/pika.md | markdown | 1.1 KB |
| references/runway.md | markdown | 1.2 KB |
| references/sora.md | markdown | 1.4 KB |
references/sora.md
View on GitHub# Sora (OpenAI adapter)
_Last verified: 2026-10-08. Verify durations, resolutions, and availability
against OpenAI's current docs before planning._
## In one line
OpenAI's video model. Known for long, coherent takes and strong
storyboard-style prompting.
## Prompt quirks
- Handles longer, more narrative prompts than most. You can brief a fuller
scene, not just a shot.
- Storyboard/timeline-style input works well: describe the sequence, not
just the frame.
- Remixing and extending clips is a core workflow: generate, then extend or
re-cut rather than re-rolling from scratch.
- Negative prompting support varies by interface; inline `Avoid:` is the
safe default.
## Hard limits
- Duration and resolution caps depend on model version and plan. Verify.
- Complex multi-character interaction still drifts; keep casts small.
## Good at / weak at
- Good at: coherent longer shots, camera motion through space, stylized
worlds.
- Weak at: exact text, precise object counts, photoreal faces in motion
(improving, but verify).
## Translating a shot-brief
- You can merge 2–3 adjacent shot-briefs into one longer Sora prompt when
the action is continuous, then split in post if needed.
- Lean on the storyboard: Sora rewards "and then" structure.
- Keep camera moves simple and motivated; it renders them cleanly.
- Still write the Preserve slot. Coherence is good, not automatic.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |