Create and edit Obsidian Flavored Markdown: wikilinks, embeds, callouts, properties, tags.
AI-pentest skill for red-teaming LLM apps: prompt injection, jailbreaks, and agentic exploitation.
OSINT methodology skill for red-team recon: domain, email, GitHub leaks, Shodan/Censys, breach data, geoint.
SQL injection testing skill: error/UNION/blind/OOB/second-order/NoSQL/GraphQL with WAF bypass and SQLmap.
Coordinate multi-agent parallel development with file ownership and interface contracts.
Implement PCI DSS controls for secure cardholder-data handling and payment systems.
Read, create, edit, merge, split, watermark, encrypt, fill forms, and OCR PDF files.
Fill any PDF locally with precise value placement in a visual browser editor.
Convert a PDF into one self-contained readable HTML file, optionally translating while keeping figures.
Structured methodology and checklist for planning and scoping penetration tests.
Optimizes app performance, Core Web Vitals, and load times by fixing profiled bottlenecks.
Pinia type-safe Vue state management: stores, state/getters/actions, store patterns.
Model-invoked Claude Code skill for general-purpose browser automation with Playwright.
Documents the .claude/plugin-name.local.md pattern for storing user-configurable Claude Code plugin settings.
Guidance for scaffolding Claude Code plugins: directory layout, plugin.json manifest, and component auto-discovery.
Generate podcast-style audio via Azure OpenAI GPT Realtime Mini over WebSocket, full-stack.
Write blameless postmortems with root-cause analysis, timelines, and action items.
Create, read, edit, and template PowerPoint .pptx slide decks and presentations.
Set up new projects (or improve existing ones) with tooling, docs and workflow automation.
Advanced prompt patterns: chain-of-thought, few-shot, structured templates for production LLMs.
Safety skill for deploying Prompt Guard to detect prompt-injection and jailbreak inputs.
Transform vague prompts into precise EARS-syntax specifications for AI code/product generation.
Configure and run LLM evaluations with Promptfoo — configs, Python assertions, llm-rubric judging.
Cedar policy enforcement and Ed25519 signed receipts gating every Claude Code tool call.