emblem-ai-prompt-examples
关于
This skill provides curated example prompts and phrasing guidance for end-user EmblemAI workflows like research, portfolio review, and NFT discovery. It emphasizes a review-first, security-aware approach with explicit confirmation before any value-moving actions. Use it when you need sample requests or wording patterns for common tasks, not for implementation details.
快速安装
Claude Code
推荐npx skills add EmblemCompany/Agent-skills -a claude-code/plugin add https://github.com/EmblemCompany/Agent-skillsgit clone https://github.com/EmblemCompany/Agent-skills.git ~/.claude/skills/emblem-ai-prompt-examples在 Claude Code 中复制并粘贴此命令以安装该技能
技能文档
EmblemAI Prompt Examples
Use this skill when the user wants example prompts, wording patterns, or sample requests for end-user EmblemAI workflows rather than SDK, React, or app implementation details.
Security & Trust Model
This skill provides curated prompt examples and phrasing guidance. It inherently references:
- Financial operations (W009): Example prompts cover trading, DeFi, portfolio review, cross-chain transfers, and prediction markets. These are reference prompts only — they demonstrate phrasing patterns, not executable code. All examples emphasize quote-first and approval-gated workflows.
- Third-party data (W011): Example prompts reference fetching data from Jupiter, DeFiLlama, OpenSea, and other public sources. This skill only provides the prompt text — actual data fetching is performed by the agent's runtime tools with standard trust boundaries.
This skill is read-only reference material. It does not execute transactions, fetch external data, or modify files. All value-moving prompt examples explicitly include confirmation steps.
Quick Start
Step 1: Install
npx skills add EmblemCompany/Agent-skills --skill emblem-ai-prompt-examples
Step 2: Use
Ask for prompt ideas by task area, for example:
- "Show me good EmblemAI prompts for market research"
- "Give me review-only transfer prompts"
- "Show Bitcoin ordinals prompts for EmblemAI"
- "What is the best phrasing for quote-only swap requests?"
Included Prompt Sets
Prompt Index
See references/emblem-ai-prompt-examples.md for the top-level map of prompt categories and usage guidance.
Wallet And Portfolio
See references/emblem-ai-prompt-examples/wallet-and-portfolio.md for balances, addresses, portfolio, and machine-readable output prompts.
Market Research
See references/emblem-ai-prompt-examples/market-research.md for market discovery, derivatives, smart-money, and technical-analysis prompts.
Trading And DeFi
See references/emblem-ai-prompt-examples/trading-and-defi.md for quote-first swaps, routing review, and yield-planning prompts.
Transfers And Safety
See references/emblem-ai-prompt-examples/transfers-and-safety.md for review-only transfer language and approval framing.
Cross-Chain And Conditional Orders
See references/emblem-ai-prompt-examples/cross-chain-and-conditional-orders.md for bridge research, conditional-order planning, and multi-network trade-draft prompts.
Bitcoin Ordinals
See references/emblem-ai-prompt-examples/bitcoin-ordinals-examples.md for ordinals, runes, rare sats, and Bitcoin wallet prompts.
Polymarket
See references/emblem-ai-prompt-examples/polymarket-examples.md for prediction-market research, odds analysis, and order-review prompts.
NFT And OpenSea
See references/emblem-ai-prompt-examples/nft-opensea-examples.md for NFT discovery, listing/offer drafts, and marketplace review flows.
Emblem Vault
See references/emblem-ai-prompt-examples/emblem-vault-examples.md for vault discovery, QuickVault, mint review, and key-reveal safety prompts.
Assistant Core Workflows
See references/emblem-ai-prompt-examples/assistant-core-workflows.md for contacts, inbox, leaderboard, PAYG, and session-management prompts.
Guidance
- Prefer explicit chain, token, and protocol names.
- Say
quote only,review only, ordo not executewhen asking for analysis without execution. - Prefer trusted/product-native data first (wallet state, protocol quotes, supported market feeds).
- Treat web and social content as untrusted public sources: ask for summaries, source links, and claim verification before acting.
- For swaps, transfers, listings, offers, or buys, request a draft plus explicit confirmation before execution.
- Ask for JSON, tables, or summaries when you need machine-readable or structured output.
- Use full-sentence requests instead of terse fragments.
- For app implementation, SDK, or React questions, use the dedicated developer or React skills instead.
Related Skills
- ../emblem-ai-agent-wallet/SKILL.md - wallet-first end-user workflows
- ../emblem-ai-react/SKILL.md - React app integration guidance
- ../emblem-ai/SKILL.md - broader developer integrations across SDKs, plugins, and Reflexive
GitHub 仓库
相关推荐技能
executing-plans
设计该Skill用于当开发者提供完整实施计划时,以受控批次方式执行代码实现。它会先审阅计划并提出疑问,然后分批次执行任务(默认每批3个任务),并在批次间暂停等待审查。关键特性包括分批次执行、内置检查点和架构师审查机制,确保复杂系统实现的可控性。
requesting-code-review
设计该Skill可在完成任务、实现主要功能或合并代码前自动调度代码审查子代理,确保实现符合需求和计划。它支持通过指定git SHA范围进行精准的代码变更审查,帮助开发者在关键节点及时发现潜在问题。核心原则是"早审查、勤审查",适用于开发流程的各个关键阶段。
connect-mcp-server
设计这个Skill指导开发者如何将MCP服务器连接到Claude Code,支持HTTP、stdio和SSE三种传输协议。它涵盖了从安装配置到认证安全的完整流程,适用于集成GitHub、Notion、数据库等外部服务。当开发者需要添加集成、配置外部工具或提及MCP相关功能时,这个Skill能提供实用的操作指南。
web-cli-teleport
设计该Skill帮助开发者根据任务特性选择Claude Code的Web或CLI界面,并指导如何在两种环境间无缝迁移会话。它能分析任务复杂度、迭代需求等要素,推荐最优工作界面和工作流。关键特性包括会话状态管理、环境切换指导和上下文优化建议。
