Feast Skills
Define feature stores, debug feature retrieval and understand Feast data flows.
Original workflows for websites, images, presentations, and everything in motion.
Define feature stores, debug feature retrieval and understand Feast data flows.
Open Plannotator on the latest rendered assistant message and use the returned annotations to revise that message or continue.
Test one gameplay question with a timeboxed greybox loop, observable playtest criteria, and a keep-or-kill decision.
Guide for working with Feast (Feature Store) — defining features, configuring feature_store.yaml, retrieving features online/offline, using the CLI, and building RAG retrieval pipelines.
Research recent discussions across Reddit, X, YouTube, Hacker News, GitHub and other sources.
Save player progress and leaderboards with guarded DataStore calls, UpdateAsync, retries, and shutdown handling.
Internals of the Feast codebase — how each component works, where the key abstractions live, and the data flow through the system.
Write and refine prompts for a specified language model, coding assistant, image tool, or video tool.
Plan academic presentations around evidence and argument structure for conference talks, seminars, thesis defenses, and research briefings.
Art direction for AI-built websites and app screens, with real imagery, separate page and app workflows, and browser inspection. The main entry is design-taste-frontend.
Use Hexclave services and CLI workflows for authentication, teams, billing, email and product analytics.
Review YouTube ideas, titles, thumbnails, hooks and pacing through attention and audience-retention methods.
Plan topics, revise tweets and threads, diagnose accounts and review growth methods, including an AI and technology focus.
Create editable PowerPoint presentations from documents, reuse presentation templates and revise existing PPTX files through a local Python workflow.
How to test and debug Feast — running targeted tests, writing unit tests for new components, debugging registry and online store issues, and inspecting live feature store state.
Turn a podcast or interview transcript into a topic-led article while preserving attribution, concrete facts, and the speaker’s voice.
Forecast univariate time series with Google TimesFM, including uncertainty estimates and optional covariates.
Human and accessibility skill for antislop. Contrast, keyboard, focus, and states for real people. Includes the contrast checker.
Review changes since the latest commit with incremental impact and blast radius analysis.
Integrate persistent memory, inspect workspaces, and recall or save conversation context.
Use an Adlerian psychology framework to examine procrastination or interrupted action when you already know what you intend to do.
Diagnose and review Terraform or OpenTofu modules, state operations, CI pipelines, and infrastructure changes with version-aware guidance.
Generate or edit raster visuals with cc-haha desktop's configured ImageGen and ImageEdit tools.
Choose physics bodies, configure collision layers and masks, detect overlaps, apply forces, and cast 2D or 3D rays.