>

>

Adapt and debug existing or new models for vLLM on Ascend NPU. Implement in /vllm-workspace/vllm and /vllm-workspace/vllm-ascend, validate via direct vllm serve from /workspace, and deliver one signed commit in the current repo.

>-

Language-agnostic coding principles for maintainability, readability, and quality. Use when implementing features, refactoring code, or reviewing code quality.

Configures Laravel Nightwatch data collection, sampling rates, filtering rules, and redaction policies. Use when setting up Nightwatch, managing data volume, protecting sensitive data (PII), or optimizing event collection for production workloads.

Direnv environment management for automatic per-project shell configuration. Use when setting up .envrc files, configuring project-specific environment variables, or integrating direnv with development tools like nix, asdf, pyenv, or nvm.

Systematic web application debugging using Chrome DevTools MCP and Playwright MCP with intelligent validation and app-specific context discovery. Use for debugging web apps, APIs, authentication flows, and UI issues.

Apply when creating or editing INFO, SPEC, IMPL, TEST, FIX documents, or STRUT plans

Use when developing applications or components using TypeScript, ensuring type safety and leveraging TypeScript features.

Build Keynote-style single-file HTML slide decks with brand-ready templates, minimal navigation, and Gemini nano banana media generation. Includes Narrative Engine integration for framework-driven deck creation with 17 proven storytelling structures and 5-agent review panel. Use when creating or editing slide decks, transforming content into presentations, or generating slide visuals.

Update existing content with fresh information and improvements.

Implement ACP checkout as an MCP server, exposing checkout operations as MCP tools. Use when building an MCP-based commerce server for AI agents that use tool-calling to complete purchases.

Creates feature dashboards in Hex Threads. Given a feature name (and optionally a PRD, Figma link, or context), discovers events, builds a metric spec, gets human approval, and instructs Hex to build charts using business context. Hex writes its own SQL — Claude provides context, table/event names, and exact segmentation logic. Use when someone asks to create, build, or generate a feature dashboard.

Scan lyrics for pronunciation risks, prevent Suno mispronunciations

Generate OrcaFlex models from templates using component assembly with

Use when evaluating business documents for structural integrity and persuasive logic.

Provides comprehensive guidance for Rspack bundler including configuration, plugins, loaders, optimization, and Webpack compatibility. Use when the user asks about Rspack, needs to configure Rspack, optimize build performance, or migrate from Webpack.

Detects performance bottlenecks in Godot projects including expensive _process functions, get_node() calls in loops, instantiations in _process, and provides optimization suggestions with Godot profiler integration

[COMMUNITY] Generate French public procurement documentation aligned with code de la commande publique, UGAP catalogue, and DINUM digital standards

'Guidez la création de contenu de modèle de cas d’utilisation pour le référentiel de plans directeurs Adobe Experience Platform. Utilisez cette compétence lors de l’ajout d’un nouveau modèle de cas d’utilisation, de la création de contenu de guide d’implémentation ou lorsque l’utilisateur mentionne l’ajout de modèles au site de plans directeurs. Gère l’ensemble du workflow : collecte des informations de modèle, génération du fichier Markdown avec la structure de modèle appropriée et mise à jour de toutes les pages de référence croisée (TOC.md, overview.md).'

Senior PHP Laravel code reviewer. Use when reviewing pull requests, examining code changes vs master branch, or when the user asks for a code review. Read-only review — never modifies code.

How to create, structure, and publish self-contained Convex components with proper isolation, exports, and dependency management

Advanced GitHub CLI workflows for PR review, CI/CD debugging, Actions management, API queries, and code search. Use when the user needs to review PRs, debug failing checks, manage workflows, search across repos, or make complex gh API calls. Not needed for basic gh commands the agent already knows.

Create MADR-format Architecture Decision Record. Use when recording an architectural decision, documenting a technical choice, or creating an ADR.

Create a simple PR from staged changes with an auto-generated commit message

Comprehensive code review workflow coordinating quality, security, performance, and documentation reviewers. 4-hour timeline for thorough multi-agent review.

Game design principles. GDD structure, balancing, player psychology,

Use when expanding a README-based dApp description into a full specification with user stories and acceptance criteria.

Terraform and OpenTofu infrastructure as code — module design, state management, multi-environment setups, remote backends, secrets management, CI/CD integration. NOT for Pulumi, CDK, Ansible,

**Owner:** QA

Bug-fixing workflow that coordinates diagnosis, test-driven reproduction, root-cause analysis, and targeted fixes. Use when the user wants to fix a bug with thorough investigation and regression testing.

Orchestrate sequential documentation audits with checkpointing and resumption.

Error prevention and best practices enforcement for agent-assisted coding. Use when writing code to catch common mistakes, enforce patterns, prevent bugs, validate inputs, handle errors, follow coding standards, avoid anti-patterns, and ensure code quality through proactive checks and guardrails.

Investigate the intended scope of a codebase topic through discussion with a human, propose a research-note outline, and only create a .memory-bank/research file after explicit human confirmation.

Use this skill whenever the user wants to design, run, or refine Cloudflare D1 schema management, migrations, and data seeding for dev/staging/production environments, especially in conjunction with Hono/Workers apps.

Create a new topic file in .notes/ with frontmatter template

Use this skill when you need to analyze code for design patterns, anti-patterns, naming conventions, and code duplication. This skill excels at identifying architectural patterns, detecting code smells, and ensuring consistency across the codebase.

scripts/

Model flexible pin assignment in JITX designs. Use when user asks about provide/require patterns, @provide.one_of or @provide.subset_of decorators, programmatic Provide, pin muxing (MCU peripherals on shared pins), DiffPair P/N polarity swapping, PCIe lane swapping or width variants, DDR4 byte/bit swapping, LPDDR5 channel swapping, hierarchical provider composition, topology (>>) on pin-assigned ports, ConstrainDiffPair or ConstrainReferenceDifference with provide/require, or flexible pin mapping for FPGAs and MCUs. Covers provide, Provide, require, all_of, one_of, subset_of, and protocol-specific pin flexibility with SI constraints.

Convert markdown planning documents to Overseer tasks via MCP codemode. Use when converting plans, specs, or design docs to trackable task hierarchies.

Standardized error handling functions for formatting, logging, and returning user-friendly error messages across the application.

Expert in Docker containerization including Dockerfile optimization, multi-stage builds, Docker Compose, and container security. Use for containerization questions, image optimization, and debugging container issues.

The functional options pattern for Go constructors and public APIs. Use when designing APIs with optional configuration, especially with 3+ parameters.

Generate content drafts when a PR is merged or significant code ships. Produces a Twitter/X thread, blog post outline, and changelog entry from PR title, changed files, and commit messages. Keywords: content, tweet, blog, changelog, announce, write about, PR merged, shipped, released, draft post, social, thread

>

Web research with Graph-of-Thoughts for fast-changing topics. Use when user requests research, analysis, investigation, or comparison requiring current information. Features hypothesis testing, source triangulation, claim verification, Red Team, self-critique, and gap analysis. Supports Quick/Standard/Deep/Exhaustive tiers. Creative Mode for cross-industry innovation.

Apply when creating or editing prompts targeting GPT-5.4. Covers output contracts, scope discipline, reasoning_effort tuning, long-context handling, ambiguity management, tool persistence, completeness and verification contracts, structured extraction, research mode with citation discipline, personality controls, coding autonomy, compaction awareness, and migration from older GPT models.

Apply CO15 Combinatorial Design to systematically explore option combinations to find optimal configurations.

Analyze your Claude Code session logs to improve prompt quality, optimize tool usage, and become a better AI-native engineer.