diff --git a/README.md b/README.md index 0dae8cc..20bc415 100644 --- a/README.md +++ b/README.md @@ -243,6 +243,7 @@ Official integrations are maintained by companies building production ready MCP - Hunter Logo **[Hunter](https://github.com/hunter-io/hunter-mcp)** - Interact with the [Hunter API](https://hunter.io) to get B2B data using natural language. - Hyperbolic Labs Logo **[Hyperbolic](https://github.com/HyperbolicLabs/hyperbolic-mcp)** - Interact with Hyperbolic's GPU cloud, enabling agents and LLMs to view and rent available GPUs, SSH into them, and run GPU-powered workloads for you. - Hyperbrowsers23 Logo **[Hyperbrowser](https://github.com/hyperbrowserai/mcp)** - [Hyperbrowser](https://www.hyperbrowser.ai/) is the next-generation platform empowering AI agents and enabling effortless, scalable browser automation. +- **[IBM watsonx.data intelligence](https://github.com/IBM/data-intelligence-mcp-server)** - Find, understand, and work with your data in the watsonx.data intelligence governance & catalog, data quality, data lineage, and data product hub - **[IBM wxflows](https://github.com/IBM/wxflows/tree/main/examples/mcp/javascript)** - Tool platform by IBM to build, test and deploy tools for any data source - Improve Digital Icon **[Improve Digital Publisher MCP](https://github.com/azerion/improvedigital-publisher-mcp-server)** - An MCP server that enables publishers to integrate [Improve Digital’s](https://improvedigital.com/) inventory management system with their AI tools or agents. - Inbox Zero Logo **[Inbox Zero](https://github.com/elie222/inbox-zero/tree/main/apps/mcp-server)** - AI personal assistant for email [Inbox Zero](https://www.getinboxzero.com) @@ -353,6 +354,7 @@ Official integrations are maintained by companies building production ready MCP - Oracle Logo **[Oracle](https://docs.oracle.com/en/database/oracle/sql-developer-command-line/25.2/sqcug/starting-and-managing-sqlcl-mcp-server.html#GUID-5F916B5D-8670-42BD-9F8B-D3D2424EC47E)** - Official [Oracle Database: SQLcl ](https://www.oracle.com/database/sqldeveloper/technologies/sqlcl/download/) MCP server enabling all access to any Oracle Database via native MCP support directly in SQLcl. - Orshot Logo **[Orshot](https://github.com/rishimohan/orshot-mcp-server)** - Official [Orshot](https://orshot.com) MCP server to dynamically generate images from custom design templates. - Oxylabs Logo **[Oxylabs](https://github.com/oxylabs/oxylabs-mcp)** - Scrape websites with Oxylabs Web API, supporting dynamic rendering and parsing for structured data extraction. +- PandaDoc **[PandaDoc](https://developers.pandadoc.com/docs/use-pandadoc-mcp-server)** - Configure AI development tools to connect to PandaDoc's Model Context Protocol server and leverage AI-powered PandaDoc integrations. - Paddle Logo **[Paddle](https://github.com/PaddleHQ/paddle-mcp-server)** - Interact with the Paddle API. Manage product catalog, billing and subscriptions, and reports. - **[PaddleOCR](https://paddlepaddle.github.io/PaddleOCR/latest/en/version3.x/deployment/mcp_server.html)** - An MCP server that brings enterprise-grade OCR and document parsing capabilities to AI applications. - PagerDuty Logo **[PagerDuty](https://github.com/PagerDuty/pagerduty-mcp-server)** - Interact with your PagerDuty account, allowing you to manage incidents, services, schedules, and more directly from your MCP-enabled client. @@ -614,6 +616,7 @@ A growing set of community-developed and maintained servers demonstrates various - **[Chroma](https://github.com/privetin/chroma)** - Vector database server for semantic document search and metadata filtering, built on Chroma - **[Chrome history](https://github.com/vincent-pli/chrome-history-mcp)** - Talk with AI about your browser history, get fun ^_^ - **[CIViC](https://github.com/QuentinCody/civic-mcp-server)** - MCP server for the Clinical Interpretation of Variants in Cancer (CIViC) database, providing access to clinical variant interpretations and genomic evidence for cancer research. +- **[claude-faf-mcp](https://github.com/Wolfe-Jam/claude-faf-mcp)** - MCP server for .faf format. Context scoring engine with project context management. - **[Claude Thread Continuity](https://github.com/peless/claude-thread-continuity)** - Persistent memory system enabling Claude Desktop conversations to resume with full context across sessions. Maintains conversation history, project states, and user preferences for seamless multi-session workflows. - **[ClaudePost](https://github.com/ZilongXue/claude-post)** - ClaudePost enables seamless email management for Gmail, offering secure features like email search, reading, and sending. - **[CLDGeminiPDF Analyzer](https://github.com/tfll37/CLDGeminiPDF-Analyzer)** - MCP server tool enabling sharing large PDF files to Google LLMs via API for further/additional analysis and response retrieval to Claude Desktop. @@ -673,6 +676,7 @@ A growing set of community-developed and maintained servers demonstrates various - **[DeepSeek MCP Server](https://github.com/DMontgomery40/deepseek-mcp-server)** - Model Context Protocol server integrating DeepSeek's advanced language models, in addition to [other useful API endpoints](https://github.com/DMontgomery40/deepseek-mcp-server?tab=readme-ov-file#features) - **[deepseek-thinker-mcp](https://github.com/ruixingshi/deepseek-thinker-mcp)** - A MCP (Model Context Protocol) provider Deepseek reasoning content to MCP-enabled AI Clients, like Claude Desktop. Supports access to Deepseek's thought processes from the Deepseek API service or from a local Ollama server. - **[Deepseek_R1](https://github.com/66julienmartin/MCP-server-Deepseek_R1)** - A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3) +- **[DeFi Rates](https://github.com/qingfeng/defi-rates-mcp)** - Query real-time DeFi lending rates across 13+ protocols (Aave, Morpho, Compound, Venus, Solend, Drift, Jupiter, etc.). Compare rates, search best opportunities, and calculate looping strategies across Ethereum, Arbitrum, Base, BSC, Solana, and HyperEVM. - **[Depyler](https://github.com/paiml/depyler/blob/main/docs/mcp-integration.md)** - Energy-efficient Python-to-Rust transpiler with progressive verification, enabling AI assistants to convert Python code to safe, performant Rust while reducing energy consumption by 75-85%. - **[deploy-mcp](https://github.com/alexpota/deploy-mcp)** - Universal deployment tracker for AI assistants with live status badges and deployment monitoring. - **[Descope](https://github.com/descope-sample-apps/descope-mcp-server)** - An MCP server to integrate with [Descope](https://descope.com) to search audit logs, manage users, and more. @@ -1016,6 +1020,7 @@ A growing set of community-developed and maintained servers demonstrates various - **[Naver](https://github.com/isnow890/naver-search-mcp)** (by isnow890) - MCP server for Naver Search API integration, supporting blog, news, shopping search and DataLab analytics features. - **[NBA](https://github.com/Taidgh-Robinson/nba-mcp-server)** - This MCP server provides tools to fetch recent and historical NBA games including basic and advanced statistics. - **[NCI GDC](https://github.com/QuentinCody/nci-gdc-mcp-server)** - Unofficial MCP server for the National Cancer Institute's Genomic Data Commons (GDC), providing access to harmonized cancer genomic and clinical data for oncology research. +- **[NCP](https://github.com/portel-dev/ncp)** (Natural Context Provider by portel.dev) - NCP lets your AI dream of a tool and articulate its need as a user story. NCP then intelligently discovers and makes that tool instantly available, streamlining thought processes, eliminating cognitive overload, and slashing token costs by up to 87% (47ms discovery). Experience true on-demand tool access, smart health monitoring, and energy efficiency for your AI agents. - **[Neo4j](https://github.com/da-okazaki/mcp-neo4j-server)** - A community built server that interacts with Neo4j Graph Database. - **[Neovim](https://github.com/bigcodegen/mcp-neovim-server)** - An MCP Server for your Neovim session. - **[Netbird](https://github.com/aantti/mcp-netbird)** - List and analyze Netbird network peers, groups, policies, and more. diff --git a/src/everything/everything.ts b/src/everything/everything.ts index 465cd51..2300ee0 100644 --- a/src/everything/everything.ts +++ b/src/everything/everything.ts @@ -6,7 +6,6 @@ import { CompleteRequestSchema, CreateMessageRequest, CreateMessageResultSchema, - ElicitRequest, ElicitResultSchema, GetPromptRequestSchema, ListPromptsRequestSchema, @@ -259,22 +258,6 @@ export const createServer = () => { }; - const requestElicitation = async ( - message: string, - requestedSchema: any, - sendRequest: SendRequest - ) => { - const request: ElicitRequest = { - method: 'elicitation/create', - params: { - message, - requestedSchema, - }, - }; - - return await sendRequest(request, ElicitResultSchema); - }; - const ALL_RESOURCES: Resource[] = Array.from({ length: 100 }, (_, i) => { const uri = `test://static/resource/${i + 1}`; if (i % 2 === 0) { @@ -552,7 +535,7 @@ export const createServer = () => { }); if (clientCapabilities!.elicitation) tools.push ({ name: ToolName.ELICITATION, - description: "Demonstrates the Elicitation feature by asking the user to provide information about their favorite color, number, and pets.", + description: "Elicitation test tool that demonstrates how to request user input with various field types (string, boolean, email, uri, date, integer, number, enum)", inputSchema: zodToJsonSchema(ElicitationSchema) as ToolInput, }); @@ -747,27 +730,75 @@ export const createServer = () => { if (name === ToolName.ELICITATION) { ElicitationSchema.parse(args); - const elicitationResult = await requestElicitation( - 'What are your favorite things?', - { - type: 'object', - properties: { - color: { type: 'string', description: 'Favorite color' }, - number: { - type: 'integer', - description: 'Favorite number', - minimum: 1, - maximum: 100, - }, - pets: { - type: 'string', - enum: ['cats', 'dogs', 'birds', 'fish', 'reptiles'], - description: 'Favorite pets', + const elicitationResult = await extra.sendRequest({ + method: 'elicitation/create', + params: { + message: 'Please provide inputs for the following fields:', + requestedSchema: { + type: 'object', + properties: { + name: { + title: 'Full Name', + type: 'string', + description: 'Your full, legal name', + }, + check: { + title: 'Agree to terms', + type: 'boolean', + description: 'A boolean check', + }, + color: { + title: 'Favorite Color', + type: 'string', + description: 'Favorite color (open text)', + default: 'blue', + }, + email: { + title: 'Email Address', + type: 'string', + format: 'email', + description: 'Your email address (will be verified, and never shared with anyone else)', + }, + homepage: { + type: 'string', + format: 'uri', + description: 'Homepage / personal site', + }, + birthdate: { + title: 'Birthdate', + type: 'string', + format: 'date', + description: 'Your date of birth (will never be shared with anyone else)', + }, + integer: { + title: 'Favorite Integer', + type: 'integer', + description: 'Your favorite integer (do not give us your phone number, pin, or other sensitive info)', + minimum: 1, + maximum: 100, + default: 42, + }, + number: { + title: 'Favorite Number', + type: 'number', + description: 'Favorite number (there are no wrong answers)', + minimum: 0, + maximum: 1000, + default: 3.14, + }, + petType: { + title: 'Pet type', + type: 'string', + enum: ['cats', 'dogs', 'birds', 'fish', 'reptiles'], + enumNames: ['Cats', 'Dogs', 'Birds', 'Fish', 'Reptiles'], + default: 'dogs', + description: 'Your favorite pet type', + }, }, + required: ['name'], }, }, - extra.sendRequest - ); + }, ElicitResultSchema, { timeout: 10 * 60 * 1000 /* 10 minutes */ }); // Handle different response actions const content = []; @@ -775,19 +806,30 @@ export const createServer = () => { if (elicitationResult.action === 'accept' && elicitationResult.content) { content.push({ type: "text", - text: `✅ User provided their favorite things!`, + text: `✅ User provided the requested information!`, }); // Only access elicitationResult.content when action is accept - const { color, number, pets } = elicitationResult.content; + const userData = elicitationResult.content; + const lines = []; + if (userData.name) lines.push(`- Name: ${userData.name}`); + if (userData.check !== undefined) lines.push(`- Agreed to terms: ${userData.check}`); + if (userData.color) lines.push(`- Favorite Color: ${userData.color}`); + if (userData.email) lines.push(`- Email: ${userData.email}`); + if (userData.homepage) lines.push(`- Homepage: ${userData.homepage}`); + if (userData.birthdate) lines.push(`- Birthdate: ${userData.birthdate}`); + if (userData.integer !== undefined) lines.push(`- Favorite Integer: ${userData.integer}`); + if (userData.number !== undefined) lines.push(`- Favorite Number: ${userData.number}`); + if (userData.petType) lines.push(`- Pet Type: ${userData.petType}`); + content.push({ type: "text", - text: `Their favorites are:\n- Color: ${color || 'not specified'}\n- Number: ${number || 'not specified'}\n- Pets: ${pets || 'not specified'}`, + text: `User inputs:\n${lines.join('\n')}`, }); } else if (elicitationResult.action === 'decline') { content.push({ type: "text", - text: `❌ User declined to provide their favorite things.`, + text: `❌ User declined to provide the requested information.`, }); } else if (elicitationResult.action === 'cancel') { content.push({