Use this skill at the end of a Google Slides deck-creation flow run by the same agent in the current session, before reporting the finished URL to the user. Catches the failure modes a creating agent reliably misses — surviving placeholder text, unswapped template imagery, layout overflow, hidden-original leaks. Trigger when the parent google-slides workflow reaches Step 9 / QA, or when the user asks "is this deck ready?" / "QA my deck" about a deck this agent itself just edited. Do NOT trigger for reviewing decks the agent did not create — without the Step 5 pre-fill snapshot the leak check silently degrades and the result is misleading.

Review image files (PNG/JPEG) against brand guidelines — score visual compliance across 8 dimensions, identify violations with exact fixes, and apply corrections via Pillow or AI regeneration. Use when reviewing ad creatives, social media images, or AI-generated visuals for brand alignment. Triggers on "check this image against brand", "review this ad image", "score this creative", "is this image on-brand", "fix brand violations on this image". For HTML/email/SMS content review, use the brand-compliance skill instead.

Investigate Vercel deployment errors using Vercel CLI. Use when user shares a Vercel deployment error, build failure, or deployment URL that needs debugging.

Execute multiple Claude Code agents in parallel using the cpo CLI tool. Use when running parallel tasks, monitoring execution, or understanding the execution workflow.

Comprehensive guide for developing WebGPU-enabled Three.js applications using TSL (Three.js Shading Language). Covers WebGPU renderer setup, TSL syntax and node materials, compute shaders, post-processing effects, and WGSL integration. Use this skill when working with Three.js WebGPU, TSL shaders, node materials, or GPU compute in Three.js.

Design safe transaction boundaries, rollback behavior, idempotency, locking, and consistency strategies for Kotlin + Spring business workflows. Use when a feature writes to the database, spans multiple repositories, publishes messages, calls external systems, suffers from partial commits or duplicate processing, or needs precise `@Transactional`, propagation, or isolation guidance.

Generate, debug, and repair Kotlin + Spring Gradle builds with minimal, compatible changes. Use when `build.gradle.kts` or `settings.gradle.kts` is failing, plugins or toolchains are incompatible, dependency management is drifting from the Spring Boot BOM, test or runtime classpaths are broken, or a Kotlin DSL patch must be safe and incremental.

Create, search, and manage Bear notes via grizzly CLI.

OpenProse VM skill pack. Activate on any `prose` command, .prose files, or OpenProse mentions; orchestrates multi-agent workflows.

Create, search, and manage Bear notes via grizzly CLI.

OpenProse VM skill pack. Activate on any `prose` command, .prose files, or OpenProse mentions; orchestrates multi-agent workflows.

Load when reviewing, refactoring, modernizing, debugging, or designing non-trivial C++ code where ownership, lifetime, RAII, concurrency, templates, interfaces, exceptions, or Core Guidelines compliance may affect correctness, safety, maintainability, or performance.

Use when experiments complete to judge what claims the results support, what they don't, and what evidence is still missing. Codex MCP evaluates results against intended claims and routes to next action (pivot, supplement, or confirm). Use after experiments finish — before writing the paper or running ablations.

Read the latest news feed results (server/data/news-results-*.json), cluster items by topic, and generate grounded research idea seeds with citations. Use when the user wants to turn their daily news into actionable ideation proposals, or when invoked by the proactive research scout (Option C).

Workflow 4: Submission rebuttal pipeline. Parses external reviews, enforces coverage and grounding, drafts a safe text-only rebuttal under venue limits, and manages follow-up rounds. Use when user says "rebuttal", "reply to reviewers", "ICML rebuttal", "OpenReview response", or wants to answer external reviews safely.

Autonomous design space exploration loop for computer architecture and EDA. Runs a program, analyzes results, tunes parameters, and iterates until objective is met or timeout. Use when user says "DSE", "design space exploration", "sweep parameters", "optimize", "find best config", or wants iterative parameter tuning.

Retrieves protein structure data from RCSB PDB, PDBe, and AlphaFold with protein disambiguation, quality assessment, and comprehensive structural profiles. Creates detailed structure reports with experimental metadata, ligand information, and download links. Use when users need protein structures, 3D models, crystallography data, or mention PDB IDs (4-character codes like 1ABC) or UniProt accessions.

Comprehensive CRISPR screen analysis for functional genomics. Analyze pooled or arrayed CRISPR screens (knockout, activation, interference) to identify essential genes, synthetic lethal interactions, and drug targets. Perform sgRNA count processing, gene-level scoring (MAGeCK, BAGEL), quality control, pathway enrichment, and drug target prioritization. Use for CRISPR screen analysis, gene essentiality studies, synthetic lethality detection, functional genomics, drug target validation, or identifying genetic vulnerabilities.

Use when the user wants to reverse-engineer an existing image ad into a reusable prompt template. Validates via Arcads — picks gpt-image-2 or Nano Banana at Phase 1. Triggers on "clone this ad as a template", "reverse engineer this ad", "turn this ad into a prompt", "extract a template", "make this ad reusable", "add to my prompt library", "study this ad and make a template". Input is an EXISTING ad image; does NOT trigger for fresh generation (use chatgpt-image-ad or nano-banana-image-ad).

Generate one or more standalone Meta image-ad creatives via ChatGPT Image 2 (gpt-image-2) through the Arcads external API. Locks the model, auto-strips platform chrome, enforces edge-safe layouts and glyph-safety inside body text. Use when the user asks for a "gpt-image-2 ad", "ChatGPT Image ad", "Image 2 ad creative", "make a static image ad with GPT", or anchors on a need for typography-heavy / dense-text / UI-mimicry ad creatives (chat threads, comparison tables, fake search results, iOS dialogs, Slack snapshots, ChatGPT-conversation ads, Apple Notes lists). Does NOT trigger on Nano Banana cues — use nano-banana-image-ad for those.