发现最适合你需求的 AI 技能
Restructure flat MEMORY.md files into a tiered memory system with an entity graph. Use when: (1) MEMORY.md is growing unwieldy or exceeds 150 lines, (2) user asks to organize/restructure/tier memory, (3) context compaction is losing important details, (4) you need structured entity lookup (people, projects, properties, contacts) instead of scanning markdown. Triggers on organize my memory, memory is too big, restructure memory, tier my memory, make memory more efficient.
Provides NodeReal MegaNode blockchain infrastructure APIs for 25+ chains including BSC, Ethereum, opBNB, Optimism, Polygon, Arbitrum, and Klaytn. Covers standard JSON-RPC endpoints, Enhanced APIs (nr_ methods for ERC-20 token balances, NFT holdings, asset transfers), MegaFuel gasless transactions via BEP-322 paymaster, Direct Route MEV protection, Debug/Trace APIs, WebSocket subscriptions, ETH Beacon Chain consensus layer, Portal API usage monitoring, API Marketplace (NFTScan, Contracts API, SPA
Systematic X (Twitter) operations skill for founders, indie developers, and tech professionals. Implements a daily Plan-Do-Check-Act closed loop with content strategy (5-pillar system), multimodal creation, thread growth playbook, engagement and community building, product promotion integration, persona development, competitor analysis, and data-driven continuous improvement. Use when managing an X account, planning content, analyzing tweet performance, engaging with community, running competiti
Send and receive messages with other AI agents using the Universal Agent Messaging protocol.
IMAP bulk email triage — pattern-based delete/archive with dry-run mode. Use when: cleaning up large email inboxes, bulk-deleting emails from specific senders or domains, archiving newsletter/digest emails, triaging email by sender domain or subject patterns. Supports IMAP STARTTLS (e.g. Proton Bridge), dry-run preview, YAML/JSON config for patterns, and processes UIDs (not sequence numbers) for reliable bulk ops.
>-
Japanese OCR via NDLOCR-Lite (National Diet Library). Trigger on 'OCR this image', '日文OCR', 'recognize Japanese text', or any request to extract text from Japanese documents, screenshots, or scanned pages. Best for printed Japanese and vertical text. Also works for English.
>
每日推送中国热榜TOP10到飞书。从微博、知乎、百度、36氪筛选最重要的新闻,智能排除明星八卦和主观评论。
Use when the task involves reading, creating, or editing `.docx` documents, especially when formatting or layout fidelity matters; prefer `python-docx` plus the bundled `scripts/render_docx.py` for visual checks.
SurrealDB Surrealism WASM extension development. Write Rust functions, compile to WASM, deploy as database modules. Part of the surreal-skills collection.
Master Android Studio IDE with debugging, profiling, refactoring, and productivity shortcuts.