feat: add video-prompt-design subagent for Veo 3 meta prompt framework#156
feat: add video-prompt-design subagent for Veo 3 meta prompt framework#156marcusquinn merged 3 commits intomainfrom
Conversation
Add structured AI video prompt engineering subagent based on the snubroot/Veo-3-Meta-Framework. Includes 7-component format (Subject, Action, Scene, Style, Dialogue, Sounds, Technical), character consistency templates, camera positioning syntax, audio hallucination prevention, and domain-specific templates. Updates AGENTS.md subagent table and progressive disclosure section. Updates README.md with new skill entry and video creation references.
Summary of ChangesHello @marcusquinn, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request significantly enhances the system's capabilities in AI video generation by integrating a dedicated subagent for advanced prompt engineering. It provides users with a robust, structured framework to create high-quality, consistent, and detailed prompts for various AI video models, thereby improving the overall output and efficiency of video content creation. Highlights
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
|
Note Other AI code review bot(s) detectedCodeRabbit has detected other AI code review bot(s) in this pull request and will avoid duplicating their findings in the review comments. This may lead to a less comprehensive review. WalkthroughThis pull request adds a new Video Prompt Design skill: a comprehensive 7-component meta-prompt guide for AI video generation, registers it in agent configs, and updates agent and public README documentation to reference the new tool. Changes
Sequence Diagram(s)(omitted — changes are documentation and registry updates without multi-component runtime control flow) Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request introduces a new video-prompt-design subagent based on the Veo 3 Meta Framework. The changes include adding the subagent's documentation, which provides a structured 7-component framework for AI video prompt engineering, and updating the AGENTS.md and README.md files to reflect this new capability. The new subagent documentation is comprehensive and well-structured. My feedback includes a couple of suggestions to improve clarity and consistency in the prompt design guidance to ensure the AI agent can interpret it as effectively as possible.
| Action: [Movements, gestures, timing, micro-expressions] | ||
| Scene: [Environment, props, lighting, weather, time of day] | ||
| Style: [Camera shot, angle, movement, colour palette, depth of field] | ||
| Dialogue: [Character]: "Speech" (Tone: descriptor) |
There was a problem hiding this comment.
The dialogue format presented here in the quick reference is slightly different from the more detailed example provided on lines 112-113 ((Character Name): "Exact dialogue here"). To ensure maximum clarity and consistency for the agent, it's best to use the same, most complete format in all examples. This reinforces the single best practice.
| Dialogue: [Character]: "Speech" (Tone: descriptor) | |
| Dialogue: (Character Name): "Speech" (Tone: descriptor) |
| ``` | ||
|
|
||
| **Critical Techniques**: | ||
| - Camera positioning: Include `(thats where the camera is)` for spatial anchoring |
There was a problem hiding this comment.
There's a small typo here and also on line 94. "thats" should be "that's". Correcting this will improve the professionalism of the documentation and any prompts generated based on it.
| - Camera positioning: Include `(thats where the camera is)` for spatial anchoring | |
| - Camera positioning: Include `(that's where the camera is)` for spatial anchoring |
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Fri Jan 23 20:42:37 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Fix all issues with AI agents
In @.agent/tools/video/video-prompt-design.md:
- Around line 94-95: Replace the typo in the camera positioning example string:
change "thats" to the contraction "that's" in the line containing "Close-up shot
with camera positioned at counter level (thats where the camera is) as the
character demonstrates the product" so the example reads "...(that's where the
camera is)..." for correct punctuation and clarity.
| "Close-up shot with camera positioned at counter level (thats where the camera is) | ||
| as the character demonstrates the product" |
There was a problem hiding this comment.
Fix minor typo in camera positioning example.
Use “that’s” to improve clarity.
✏️ Proposed fix
-"Close-up shot with camera positioned at counter level (thats where the camera is)
+"Close-up shot with camera positioned at counter level (that's where the camera is)📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| "Close-up shot with camera positioned at counter level (thats where the camera is) | |
| as the character demonstrates the product" | |
| "Close-up shot with camera positioned at counter level (that's where the camera is) | |
| as the character demonstrates the product" |
🤖 Prompt for AI Agents
In @.agent/tools/video/video-prompt-design.md around lines 94 - 95, Replace the
typo in the camera positioning example string: change "thats" to the contraction
"that's" in the line containing "Close-up shot with camera positioned at counter
level (thats where the camera is) as the character demonstrates the product" so
the example reads "...(that's where the camera is)..." for correct punctuation
and clarity.
🤖 Augment PR SummarySummary: Introduces a
upstream_url reference and enables webfetch; the guidance standardizes “Subject/Action/Scene/Style/Dialogue/Sounds/Technical” output with explicit audio + negative constraints.
Mentions Veo 3 constraints (8s/16:9) and recommends short dialogue limits per generation.
Includes templates for consistent characters and domain-specific audio libraries to reduce hallucinations.
No runtime code changes; updates are documentation/agent-instructions only.
🤖 Was this summary useful? React with 👍 or 👎 |
![augmentcode[bot]](https://avatars.githubusercontent.com/in/1027498?s=60&v=4){kind=small}
| upstream_url: https://github.com/snubroot/Veo-3-Meta-Framework | ||
| tools: | ||
| read: true | ||
| write: true |
There was a problem hiding this comment.
video-prompt-design enables write/edit tools in frontmatter; since this subagent appears to be guidance-only, consider disabling them to follow least-privilege and reduce accidental workspace modification risk.
🤖 Was this useful? React with 👍 or 👎
| |-------|--------|-------------| | ||
| | **cloudflare-platform** | [dmmulroy/cloudflare-skill](https://github.com/dmmulroy/cloudflare-skill) | 60 Cloudflare products: Workers, Pages, D1, R2, KV, Durable Objects, AI, networking, security | | ||
| | **remotion** | [remotion-dev/skills](https://github.com/remotion-dev/skills) | Programmatic video creation with React, animations, rendering | | ||
| | **video-prompt-design** | [snubroot/Veo-3-Meta-Framework](https://github.com/snubroot/Veo-3-Meta-Framework) | AI video prompt engineering - 7-component meta prompt framework for Veo 3 | |
There was a problem hiding this comment.
This is listed as an imported skill with “automatic update tracking”, but there’s no corresponding entry for video-prompt-design in .agent/configs/skill-sources.json, so aidevops skill check/update won’t be able to track upstream changes.
🤖 Was this useful? React with 👍 or 👎
|
Re: |
…ue format, add skill tracking - Set write/edit to false (guidance-only subagent, least-privilege) - Use (Character Name): format consistently in quick reference - Add video-prompt-design entry to skill-sources.json for upstream tracking
Remove quality tiers, domain templates, meta prompt generation layers, and success metrics - these are aspirational fluff from upstream that don't help craft better prompts. Consolidate physics/movement into one section. Keep only actionable prompt engineering guidance.
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Fri Jan 23 21:19:24 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
|
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Fri Jan 23 21:20:17 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
1 participant
Summary
video-prompt-designsubagent totools/video/based on the Veo 3 Meta FrameworkChanges
.agent/tools/video/video-prompt-design.md- Complete subagent with character consistency templates, camera positioning syntax, audio hallucination prevention, domain-specific templates, and Veo 3 limitations.agent/AGENTS.md- Added to subagent folder table and progressive disclosure sectionREADME.md- Added to imported skills table, Video Creation section, and Animation & Video sectionDesign Decisions
upstream_urlin frontmatter to track source repowebfetch: truefor fetching latest model documentationtextlanguage specifier (no bare fences)Summary by CodeRabbit
✏️ Tip: You can customize this high-level summary in your review settings.