自然語言處理

Use when design is complete and you need detailed implementation tasks for engineers with zero codebase context - creates comprehensive implementation plans with exact file paths, complete code examples, and verification steps assuming engineer has minimal domain knowledge

Generate high-quality images from text prompts using fal.ai's text-to-image models. Supports intelligent model selection, style transfer, and professional-grade outputs.

Track complex, multi-session work with dependency graphs using bd (beads) issue tracker. Use when work spans multiple sessions, has complex dependencies, or requires persistent context across compaction cycles. For simple single-session linear tasks, TodoWrite remains appropriate.

Dramatic animated effects for landing pages and marketing sites. Use when user needs "wow-factor", hero sections with spotlights/aurora, 3D hover cards, text reveal animations, animated backgrounds. Built on Framer Motion + Tailwind. NOT for basic UI (use shadcn) or SaaS dashboards (use Magic UI).

Compress and simplify prompts to preserve meaning while reducing use of context

Use when validating or diagnosing the cortex environment (consistency, duplicates, redundancy, optimization) - provides a diagnostic workflow and reporting structure for context health checks.

Kubernetes and OpenShift security assessment, hardening, and compliance. Use this skill when: (1) Auditing cluster or workload security posture (2) Implementing Pod Security Standards/Admission (3) Configuring RBAC roles and permissions (4) Setting up NetworkPolicies for zero-trust (5) Managing Secrets securely (encryption, external secrets) (6) Scanning images for vulnerabilities (7) Implementing OCP SecurityContextConstraints (8) Compliance checking (CIS benchmarks, SOC2, PCI-DSS) (9) Security incident investigation (10) Hardening cluster components

Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.

Analyze complete email thread with conversation context, timeline, participants, and attachments tracking.

Synchronization patterns for CLAUDE.md and GEMINI.md memory files. Covers import syntax, drift detection, and one-way sync. Use when setting up GEMINI.md, detecting context drift between memory files, understanding @import syntax, or troubleshooting sync issues.

Inspect GitHub PR checks with gh, pull failing GitHub Actions logs, summarize failure context, then create a fix plan and implement after user approval. Use when a user asks to debug or fix failing PR CI/CD checks on GitHub Actions and wants a plan + code changes; for external checks (e.g., Buildkite), only report the details URL and mark them out of scope.

Execute GitHub operations (PRs, issues, labels, comments, merges) using PowerShell scripts with structured output and error handling. Use when working with pull requests, issues, review comments, or CI checks instead of raw gh commands.

Universal context management and planning system. PROACTIVELY activate for: (1) ANY complex task requiring planning, (2) Multi-file projects/websites/apps, (3) Architecture decisions, (4) Research tasks, (5) Refactoring, (6) Long coding sessions, (7) Tasks with 3+ sequential steps. Provides: optimal file creation order, context-efficient workflows, extended thinking delegation (23x context efficiency), passive deep analysis architecture, progressive task decomposition, and prevents redundant work. Saves 62% context on average. Essential for maintaining session performance and analytical depth.

AI-led stakeholder interviews using LLMREI research-backed patterns. Conducts structured interviews to elicit requirements through context-adaptive questioning, active listening, and systematic requirement extraction.

Audit current context composition and identify optimization opportunities. Use when context window is overloaded, agents are underperforming, or applying the R&D framework to optimize token usage.

Provides context about how worlds (games) are created, loaded, saved, shared, and fetched in Codako. Use when working on API routes for worlds, the explore page, editor save/load logic, forking, or localStorage handling for anonymous users.

Generate images using Google's Nano Banana Pro (gemini-3-pro-image-preview). Accepts text prompts and optionally images (for editing/transformation) as INPUT. Returns generated IMAGES as OUTPUT. Use when user asks to create, generate, edit, or draw images, infographics, visualizations, diagrams, charts, or illustrations. Excellent for data-accurate infographics and text rendering.

Model Context Protocol (MCP) server patterns for building integrations with Claude Code. Triggers on: mcp server, model context protocol, tool handler, mcp resource, mcp tool.

Tenant context resolution and propagation patterns for multi-tenant applications. Covers middleware, headers, claims, and distributed tracing.

Generates and maintains codebase context documentation (.context.md files). Implements the Codebase Context Specification for AI-friendly project documentation. Activates when documenting architecture or onboarding.