Auto Review closeout. Codex review is the default when no engine is set and is the recommended reviewer.
EnglishPrepare accepted storyboard frames for AI-video upload batches. Use when the user needs Omni/Gemini 10-image folders, Higgsfield 9-image groups, sequential copies, overlap frames, manifests, contact sheets, or prompt placeholders. File workflow only; does not creatively edit or generate frames.
EnglishLive watcher mode for product/dev sweeps. Use when the user asks an agent to watch an app, tail logs, verify APIs/DB state, capture experience notes, track discrepancies, or act as a scribe/observer while they test a flow.
EnglishEnter codebase mastery mode before implementation. Use when the user asks to DEVOUR a repo or codebase, learn everything, deeply onboard, map architecture, trace runtime flows, understand blast radius, or build intuition about where and when things happen in a system.
EnglishReview generated storyboard frames or contact sheets for product-launch motion quality. Use when the user asks whether frames look professional, cohesive, AI-video-ready, faithful to screenshots, or should be regenerated. Outputs keep/regenerate decisions and targeted regeneration notes; does not manually edit frames.
EnglishReason about a whole system by mapping it onto a mature, structurally-similar domain that already paid for its mistakes, then read that domain's laws, invariants, and blindspots back onto the system. Use when the user wants the big-picture / first-principles view of an architecture, codebase, product, or vision; asks "what is this really like", "find the analogy", "what are we blind to", "think about this systemically / from first principles", "is this like X"; or wants to pressure-test a design by inheriting a proven domain's failure modes. A thinking mode, not a file-producing workflow.
EnglishShape a product-launch or SaaS motion-graphics video into a tight narrative spine before storyboard frame generation. Use when the user is deciding the hook, flow, duration, voiceover, feature emphasis, or ad rhythm for a launch film. Produces beats and frame direction, not generated images.
EnglishTurn vague feedback, screenshots, taste, frustration, or half-formed product intuition into an implementation-ready prompt for a coding agent. Use for UI/UX bugs, frontend polish, animation behavior, component architecture, and "it feels wrong but I don't know how to say it" moments.
EnglishCapture a conversation's essence into the active host's memory directory. Supports Codex with a Workbench project-memory convention and Claude Code project memory without assuming the two plugin runtimes share state.
EnglishCreate AI-video motion handoff prompts and keyframe grouping plans from accepted storyboard frames. Use when the user has storyboard images and needs Higgsfield Canvas, Gemini/Omni, Runway, Krea, or another video model to understand frame order, motion intent, clip stitching, and UI stability constraints. Does not generate storyboard frames; pair with videos:storyboard only when new frames are still needed.
EnglishSee what a codebase wants to become. The generative counterpart to a foundation audit - read the existing structure and surface the features it already implies, the capacity it isn't using, and the doors it hasn't opened. Two modes - open ("what does this want to become?") and wish (the user brings "I wish it could X" and the structure answers whether and how it can grant it). Use when the user asks what could this become, squeeze the potential of this feature, what features want to be born, what's latent here, run potential, or starts a sentence with "I wish it could". A conversational thinking mode like isomorph - it never implements and never writes files.
EnglishThe reality-check hotline. Mid-conversation, leave the closed room and go to the internet to ground what was just said against how the world actually does it. Two faces - verify (we concluded something confidently; pull the load-bearing claims and check each against real fetched sources - confirmed / wrong / oversimplified / outdated / no-consensus) and discover (we're stuck or starting fresh; find the dominant pattern and the gotchas). Use when the user says "scour this", "are we on the right page?", "is this actually how people do it?", "go check the internet", "make sure we're not confidently wrong", or asks how something is normally done. Fast and conversational like isomorph - never writes files, never answers from memory.
EnglishIntelligent skill recommendation engine — analyzes your current state across all data sources (history, pipeline, expiry, performance, calendar, DNA) and recommends exactly which skills to run next, in what order, and why. The "what should I do now?" answer for 89 skills.
EnglishDistill screenshots, brand assets, and motion references into a reusable visual style contract for storyboard or launch-video generation. Use when the user is sensitive about taste, cohesion, product-launch motion-graphic vibe, reference borrowing, logo fidelity, UI realism, or avoiding generic AI visuals.
EnglishReconstruct day-by-day work trails from Codex and Claude conversation evidence into a durable agenda/domain graph. Use when the user asks to log what they did across dates, scan conversation history, update an Obsidian-style trail, or infer agendas from user messages.
EnglishIncludes Autoreview, Batch Prep, Launch Narrative, and other workflows.
Supported by the Atlas install prompt. Skill-specific requirements vary.
Pack stats
What’s included
Autoreview
Auto Review closeout. Codex review is the default when no engine is set and is the recommended reviewer.
Batch Prep
Prepare accepted storyboard frames for AI-video upload batches. Use when the user needs Omni/Gemini 10-image folders, Higgsfield 9-image groups, sequential copies, overlap frames, manifests, contact sheets, or prompt placeholders. File workflow only; does not creatively edit or generate frames.
Bedrock
AWS Bedrock foundation models for generative AI. Use when invoking foundation models, building AI applications, creating embeddings, configuring model access, or implemen
Birdwatch
Live watcher mode for product/dev sweeps. Use when the user asks an agent to watch an app, tail logs, verify APIs/DB state, capture experience notes, track discrepancies, or act as a scribe/observer while they test a flow.
Devour
Enter codebase mastery mode before implementation. Use when the user asks to DEVOUR a repo or codebase, learn everything, deeply onboard, map architecture, trace runtime flows, understand blast radius, or build intuition about where and when things happen in a system.
Frame QA
Review generated storyboard frames or contact sheets for product-launch motion quality. Use when the user asks whether frames look professional, cohesive, AI-video-ready, faithful to screenshots, or should be regenerated. Outputs keep/regenerate decisions and targeted regeneration notes; does not manually edit frames.
Isomorph
Reason about a whole system by mapping it onto a mature, structurally-similar domain that already paid for its mistakes, then read that domain's laws, invariants, and blindspots back onto the system. Use when the user wants the big-picture / first-principles view of an architecture, codebase, product, or vision; asks "what is this really like", "find the analogy", "what are we blind to", "think about this systemically / from first principles", "is this like X"; or wants to pressure-test a design by inheriting a proven domain's failure modes. A thinking mode, not a file-producing workflow.
Launch Narrative
Shape a product-launch or SaaS motion-graphics video into a tight narrative spine before storyboard frame generation. Use when the user is deciding the hook, flow, duration, voiceover, feature emphasis, or ad rhythm for a launch film. Produces beats and frame direction, not generated images.
Max Prompt
Turn vague feedback, screenshots, taste, frustration, or half-formed product intuition into an implementation-ready prompt for a coding agent. Use for UI/UX bugs, frontend polish, animation behavior, component architecture, and "it feels wrong but I don't know how to say it" moments.
Memory Scriber
Capture a conversation's essence into the active host's memory directory. Supports Codex with a Workbench project-memory convention and Claude Code project memory without assuming the two plugin runtimes share state.
Motion Handoff
Create AI-video motion handoff prompts and keyframe grouping plans from accepted storyboard frames. Use when the user has storyboard images and needs Higgsfield Canvas, Gemini/Omni, Runway, Krea, or another video model to understand frame order, motion intent, clip stitching, and UI stability constraints. Does not generate storyboard frames; pair with videos:storyboard only when new frames are still needed.
Potential
See what a codebase wants to become. The generative counterpart to a foundation audit - read the existing structure and surface the features it already implies, the capacity it isn't using, and the doors it hasn't opened. Two modes - open ("what does this want to become?") and wish (the user brings "I wish it could X" and the structure answers whether and how it can grant it). Use when the user asks what could this become, squeeze the potential of this feature, what features want to be born, what's latent here, run potential, or starts a sentence with "I wish it could". A conversational thinking mode like isomorph - it never implements and never writes files.
Scour
The reality-check hotline. Mid-conversation, leave the closed room and go to the internet to ground what was just said against how the world actually does it. Two faces - verify (we concluded something confidently; pull the load-bearing claims and check each against real fetched sources - confirmed / wrong / oversimplified / outdated / no-consensus) and discover (we're stuck or starting fresh; find the dominant pattern and the gotchas). Use when the user says "scour this", "are we on the right page?", "is this actually how people do it?", "go check the internet", "make sure we're not confidently wrong", or asks how something is normally done. Fast and conversational like isomorph - never writes files, never answers from memory.
Skill Advisor
Intelligent skill recommendation engine — analyzes your current state across all data sources (history, pipeline, expiry, performance, calendar, DNA) and recommends exactly which skills to run next, in what order, and why. The "what should I do now?" answer for 89 skills.
Style Contract
Distill screenshots, brand assets, and motion references into a reusable visual style contract for storyboard or launch-video generation. Use when the user is sensitive about taste, cohesion, product-launch motion-graphic vibe, reference borrowing, logo fidelity, UI realism, or avoiding generic AI visuals.
Trail Scriber
Reconstruct day-by-day work trails from Codex and Claude conversation evidence into a durable agenda/domain graph. Use when the user asks to log what they did across dates, scan conversation history, update an Obsidian-style trail, or infer agendas from user messages.
Quick start
- Choose your skills. Select individual skills above, or choose the entire pack.
- Copy the install prompt. Use Copy install prompt above the list.
- Send it to your agent. Paste into your coding agent’s chat. Your agent reviews the source and guides the setup.
Try it with your agent
After installation, start with a request like:
Use the Autoreview skill to help with my task.
Ask me for the context you need before getting started.Requirements & license
- A coding agent that can access your project files and install skills.
- Your agent checks each skill’s instructions, dependencies, and permissions during setup.
- License details are not provided in these listings. Check the author’s terms before using the skills.