safe-refactor
Refactor non-upgradeable Solidity contracts for clarity and maintainability without changing observable behavior, interfaces, or tests. Use when behavior is locked and you need structural cleanup.
Explore finance and payments skills for AI agents. Connect billing, subscriptions, and transaction workflows directly into Claude Code, Cursor, and automation agents.
This directory brings installable AI Agent skills into one place so you can filter by search, category, topic, and official source, then install them directly into Claude Code, Cursor, Windsurf, and other supported environments.
Refactor non-upgradeable Solidity contracts for clarity and maintainability without changing observable behavior, interfaces, or tests. Use when behavior is locked and you need structural cleanup.
Generate marketing content tailored for clinical trial EDC audiences including blog posts, thought leadership articles, and social media content with industry-appropriate tone and SEO optimization.
When the user needs marketing ideas, inspiration, or strategies for their SaaS or software product. Also use when the user asks for marketing ideas, growth ideas, how to market, marketing strategies, marketing tactics, ways to promote, or ideas to grow. This skill provides 139 proven marketing approaches organized by category.
Use when debugging failed CI/CD jobs, analyzing build logs, or investigating GitHub Actions failures. Access BuildBuddy remote build execution and caching service for detailed build insights.
Error handling patterns for stock_picker_bot using FastAPI exception handlers, try/except blocks, and logging. Covers API errors, database failures, yfinance failures, and LLM API errors.
Build, refactor, and review Stratify platform features while enforcing cache-first market-data flows, tab preloading, terminal-pro dark UI patterns, Vercel serverless API architecture, and production delivery conventions. Use when work touches Stratify dashboard components, API routes, strategy builder behavior, market data streaming, deployment readiness, or AI search result rendering.
Search and discover Claude Code skills and MCP servers from marketplaces, GitHub repositories, and registries. Use when (1) user asks to find skills for a specific task, (2) looking for MCP servers to connect external tools, (3) user mentions find skill, search MCP, discover tools, or what skills exist for X, (4) before creating a custom skill to check if one already exists.
Product and delivery orchestration workflow for an AI or agent development team building features from specs, scope docs, and an existing codebase. Use when Codex needs to coordinate multiple agents in parallel without conflicts, turn requirements into execution plans, define ownership and integration rules, cut backlog slices, manage dependencies, write acceptance criteria, or produce delivery status and blocker escalations.
AI-assisted development project template with Claude Skills, Memory Bank, and Constitution-Bylaw system
Distributed adversarial behavioral security evaluation framework for LLMs - Swarm-based parallel probing with cryptographic consensus
Route tasks to agents using capability scoring, detect stalls, and manage dependency cascades. Use when assigning work to agents, investigating stalled tasks, or managing multi-agent coordination.
arxiv-search - это агент поиска исследовательских статей, использующий TypeScript и Claude Agent SDK ИИ Агенты