尚未偵測到更新。GitHub 發布版本後會顯示在這裡。
個人 GitHub 帳號下公開儲存庫的擁有者可以登入並驗證所有權,自行管理介紹,無需管理員核准。
登入並產生驗證檔案,在 30 分鐘內將其提交至此儲存庫的預設分支。其他分支或分叉儲存庫中的檔案無效。然後在下方檢查檔案,即可無需管理員核准直接管理專案。請保留檔案以維持管理權限。
如需登入,將前往登入頁面。登入後請回到此處,再次點擊所選的驗證按鈕。
在專案資料夾中執行 /nomorevibe verify,即可驗證所有權並自行管理介紹。
一分鐘完成驗證可安裝的 Hive Intelligence 軟體代理技能,適用於加密 MCP 設定、發現與研究工作流程。
AI 翻譯
Installable Hive Intelligence agent skills for crypto MCP setup, discovery, and research workflows.
取自儲存庫 README
正在載入完整 README。 以下為已儲存的摘錄。
Hive Skills: Agent Skills for Crypto Intelligence
Live crypto, wallet, token, DeFi, DEX, NFT, Solana, security, network, prediction-market, and market workflows for Claude Code, Cursor, Codex, and any agent, powered by the Hive Intelligence MCP server.
Install a single skill, non-interactively, or for a specific agent:
Skills are also bundled in the hive-intelligence npm package (npx -y -p hive-intelligence@latest hive init --all --browser) and installable as a Claude plugin via the bundled .claude-plugin/plugin.json.檢視完整 README ↗Live crypto, wallet, token, DeFi, DEX, NFT, Solana, security, network, prediction-market, and market workflows for Claude Code, Cursor, Codex, and any agent, powered by the Hive Intelligence MCP server.
npx skills add hive-intel/hive-skills # all skills
Install a single skill, non-interactively, or for a specific agent:
npx skills add hive-intel/hive-skills --skill hive-token-diligence --yes
npx skills add hive-intel/hive-skills --skill hive-query --agent claude-code
npx skills add hive-intel/hive-skills --list # browse without installing
Skills are also bundled in the hive-intelligence npm package (npx -y -p hive-intelligence@latest hive init --all --browser) and installable as a Claude plugin via the bundled .claude-plugin/plugin.json.
https://mcp.hiveintelligence.xyz/mcp. Works with a single Hive API key and no provider setup. Start with the hive-mcp skill to wire it up, or hive-build-onboarding to get a key.hive-mcp skill covers both paths.The one skill to install first: live crypto intelligence on MCP or the hive CLI, with setup, routing, credits, error handling, and feedback built in.
| Skill | Use when |
|---|---|
hive | One skill for live crypto intelligence on MCP or the hive CLI: setup, routing, the Hive loop, credits and errors, feedback, and archived derivatives history. |
Skills for installing Hive MCP, getting API keys, using the Hive CLI, and integrating Hive into application code.
| Skill | Use when |
|---|---|
hive-mcp | Add Hive's MCP server to Claude Code, Cursor, VS Code, Windsurf, Claude Desktop, ChatGPT, or Gemini CLI. Per-client instructions. |
hive-build-onboarding | Walk the user through browser-based sign-in. Use when the user has no API key or cannot find theirs. |
hive-cli | Use the hive CLI inline from a chat: query prices, scan wallets, check token security, automate briefings. |
hive-build | Integrate Hive into app code via REST or MCP SDK. Python, TypeScript, Go, Java, Rust patterns with retry / async / typed responses. |
Skills that teach agents how to discover task toolsets, inspect schemas, and execute bounded Hive MCP calls.
| Skill | Use when |
|---|---|
hive-query | Route crypto questions through canonical task toolsets, schema lookup, and exact tool invocation. |
hive-tool-discovery | Find the right Hive tool, task toolset, schema, or endpoint before execution. |
Focused workflows for market, token, wallet, security, DEX, DeFi, NFT, Solana, network, prediction-market, and stateful monitoring analysis.
| Skill | Use when |
|---|---|
hive-market-research | Research live prices, liquidity, exchanges, OHLC, order books, funding rates, and derivatives. |
hive-token-diligence | Investigate token metadata, market context, liquidity, holders, enrichment, and risk. |
hive-wallet-investigation | Inspect wallet balances, transfers, NFTs, PnL, DeFi positions, and notable activity. |
hive-security-risk | Check token, approval, address, dApp, phishing, contract, and simulation risk. |
hive-dex-pool-analysis | Analyze pools, pairs, liquidity, trades, OHLCV, and DEX flow. |
hive-defi-research | Analyze DeFi protocols, TVL, fees, revenue, stablecoins, bridges, and yields. |
hive-nft-research | Research NFT collections, ownership, metadata, floors, sales, rarity, and spam checks. |
hive-solana-analysis | Analyze Solana wallets, SPL accounts, DAS assets, parsed transactions, and priority fees. |
hive-network-infrastructure | Read chain state, gas, blocks, receipts, logs, transaction status, and RPC diagnostics. |
hive-prediction-markets | Research Polymarket prediction markets: events, market-implied odds, order books, odds history, and wallet positions. |
hive-stateful-monitoring | Create, list, update, and archive durable crypto monitors, alerts, scheduled reports, and agent memory. |
Once installed, skills activate when the agent detects a relevant task. Each example shows the phrasing that routes to a given skill:
| Ask your agent | Routes to |
|---|---|
| "What is the price and 24h change for ETH right now?" | hive-query |
| "Is the token at 0x6982… safe to ape? Run diligence." | hive-token-diligence |
| "Trace where this wallet's funds came from on Ethereum." | hive-wallet-investigation |
| "What's the BTC perp funding-rate picture across exchanges?" | hive-market-research |
| "Show the top DeFi protocols by TVL and 7d fees." | hive-defi-research |
| "Add Hive to Cursor and verify it works." | hive-mcp |
Answers should report provider provenance, data freshness, and runtime status.
Each skill follows the Agent Skills spec: a SKILL.md with YAML name + description and progressive disclosure: metadata loads at startup, full instructions load on activation, and where a skill carries genuinely deep material (per-client install matrices, root-MCP architecture, B2B subject signing) it lives in references/ and loads only when needed. Every skill also ships evals/ with execution test cases and trigger evals. Every Hive skill preserves the same loop: discovery → schema lookup → bounded execution → diagnostics → provenance-aware answer.
Data trust is the product. Skills surface the source provider, freshness, and degraded/cached/rate-limited runtime states in grounded answers, and avoid silently mixing provider data. Missing provider keys return a classified missing_key state rather than a wrong answer.
Claude Desktop, Claude Code, Cursor, Windsurf, VS Code (Copilot Chat), OpenAI Responses API, Codex CLI, Gemini CLI, and agents installable via npx skills or a .claude-plugin marketplace.
This repository is an auto-generated mirror of the Hive monorepo. Report issues or request changes via GitHub Issues or [email protected]. Upstream fixes sync back here.
https://mcp.hiveintelligence.xyz/mcpMIT. See LICENSE. Skills are MIT licensed; the Hive MCP server itself is a hosted product.