{"_id":"@agentai/mcp-server","_rev":"3-056772ea0f365df9d903ef6bee68118b","name":"@agentai/mcp-server","dist-tags":{"latest":"0.1.1"},"versions":{"0.1.0":{"name":"@agentai/mcp-server","version":"0.1.0","author":{"name":"Andrei Oprisan","email":"andrei@agent.ai"},"license":"MIT","_id":"@agentai/mcp-server@0.1.0","maintainers":[{"name":"andreioprisan","email":"andrei@oprisan.com"}],"homepage":"https://docs.agent.ai","bugs":{"url":"https://github.com/OnStartups/agentai-mcp-server/servers/issues"},"bin":{"mcp-server-agentai":"index.js"},"dist":{"shasum":"4b57859910906047e83e774a2237aa8649bedb37","tarball":"https://registry.npmjs.org/@agentai/mcp-server/-/mcp-server-0.1.0.tgz","fileCount":3,"integrity":"sha512-xXv/ksDpUfRkD2aMhRV8FufA7NSkRwcqTEvofPcfoy+Fvd8LUkMBdyuWBL/kLYiCHLLPS2r6D6aLKhpXWPoVpg==","signatures":[{"sig":"MEYCIQCH/N+5CoksJZdraRhRxX03bc8ARscErYNQJGdt0BEd1AIhAOzhQ6yb8igVag8DHCJOaMK4ix57ZHn94hE78cMwtfXn","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":9853},"type":"module","gitHead":"39d99875928aec40978840eeca5cd506e0e35d62","scripts":{"start":"node index.js","prepare":"shx chmod +x index.js"},"_npmUser":{"name":"andreioprisan","email":"andrei@oprisan.com"},"_npmVersion":"9.6.7","description":"MCP server for agent.ai API integration","directories":{},"_nodeVersion":"18.17.1","dependencies":{"@modelcontextprotocol/sdk":"1.0.1"},"_hasShrinkwrap":false,"devDependencies":{"shx":"^0.3.4"},"_npmOperationalInternal":{"tmp":"tmp/mcp-server_0.1.0_1741472390323_0.035034270647698706","host":"s3://npm-registry-packages-npm-production"}},"0.1.1":{"name":"@agentai/mcp-server","version":"0.1.1","description":"MCP server for agent.ai API integration","license":"MIT","author":{"name":"Andrei Oprisan","email":"andrei@agent.ai"},"homepage":"https://docs.agent.ai","bugs":{"url":"https://github.com/OnStartups/agentai-mcp-server/servers/issues"},"type":"module","bin":{"mcp-server-agentai":"index.js"},"scripts":{"prepare":"shx chmod +x index.js","start":"node index.js"},"dependencies":{"@agentai/mcp-server":"^0.1.0","@modelcontextprotocol/sdk":"1.0.1"},"devDependencies":{"shx":"^0.3.4"},"gitHead":"39d99875928aec40978840eeca5cd506e0e35d62","_id":"@agentai/mcp-server@0.1.1","_nodeVersion":"18.17.1","_npmVersion":"9.6.7","dist":{"integrity":"sha512-UH6KCeAUO7uKc61WMFn/tpvZvWGUQ4V9vnRLYd7KH9H5fdNtFvOo8Cu3QCyOTN2hbbgyWHTZiLQESfSqbH1MYA==","shasum":"4bffe0f126d4b60f67e31c97fd6707e5898af464","tarball":"https://registry.npmjs.org/@agentai/mcp-server/-/mcp-server-0.1.1.tgz","fileCount":3,"unpackedSize":10258,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEQCIBy7rgzgVeIUeevjxY5UVCbrFH8LWdDjtC49ePSpPYrqAiBRLk7kpwL7ajNtBXDk8p+WAF236aJuvYaEjX3m6tZhgQ=="}]},"_npmUser":{"name":"andreioprisan","email":"andrei@oprisan.com"},"directories":{},"maintainers":[{"name":"andreioprisan","email":"andrei@oprisan.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/mcp-server_0.1.1_1743376826733_0.7858238413332077"},"_hasShrinkwrap":false}},"time":{"created":"2025-03-08T22:19:50.322Z","modified":"2025-03-30T23:20:27.071Z","0.1.0":"2025-03-08T22:19:50.516Z","0.1.1":"2025-03-30T23:20:26.886Z"},"bugs":{"url":"https://github.com/OnStartups/agentai-mcp-server/servers/issues"},"author":{"name":"Andrei Oprisan","email":"andrei@agent.ai"},"license":"MIT","homepage":"https://docs.agent.ai","description":"MCP server for agent.ai API integration","maintainers":[{"name":"andreioprisan","email":"andrei@oprisan.com"}],"readme":"# Agent.ai MCP Server\n\nAn MCP server implementation that integrates with the Agent.ai API, providing web text extraction, web screenshots, and YouTube transcript capabilities through a dynamic function loading system.\n\n## Features\n\n- **Dynamic Function Loading**: Automatically fetches available functions from Agent.ai API\n- **Web Text Extraction**: Scrape or crawl web pages for text content\n- **Web Screenshots**: Capture visual screenshots of web pages\n- **YouTube Transcripts**: Extract transcripts from YouTube videos\n- **Caching**: Efficient caching of function definitions to reduce API calls\n\n## Tools\n\nThe server dynamically loads tools from the Agent.ai API. The currently available tools include:\n\n- **grab_web_text**\n  - Extract text content from web pages\n  - Inputs:\n    - `url` (string, required): URL of the web page to extract\n    - `mode` (string, optional): \"scrape\" for one page, \"crawl\" for up to 100 pages\n  \n- **grab_web_screenshot**\n  - Capture visual screenshots of web pages\n  - Inputs:\n    - `url` (string, required): URL of the web page to capture\n    - `ttl_for_screenshot` (integer, optional): Cache expiration time in seconds\n\n- **get_youtube_transcript**\n  - Fetch transcripts from YouTube videos\n  - Inputs:\n    - `url` (string, required): URL of the YouTube video\n\nand dozens of other tools. To see all available tools, visit https://docs.agent.ai/api-reference.\n\n## Configuration\n\n### Getting an API Token\nTo use this MCP server, you'll need an Agent.ai API token. Contact Agent.ai to obtain your token.\n\n### Usage with Claude Desktop\nAdd this to your `claude_desktop_config.json`:\n\n### Docker\n\n```json\n{\n  \"mcpServers\": {\n    \"agentai\": {\n      \"command\": \"docker\",\n      \"args\": [\n        \"run\",\n        \"-i\",\n        \"--rm\",\n        \"-e\",\n        \"API_TOKEN\",\n        \"mcp/agentai\"\n      ],\n      \"env\": {\n        \"API_TOKEN\": \"YOUR_API_TOKEN_HERE\",\n        \"INCLUDE_AGENTS\": \"YOUR_AGENTS_LISTING_LOGIC_HERE\"\n      }\n    }\n  }\n}\n```\n\n### NPX\n\n```json\n{\n  \"mcpServers\": {\n    \"agentai\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"@agentai/mcp-server\"\n      ],\n      \"env\": {\n        \"API_TOKEN\": \"YOUR_API_TOKEN_HERE\",\n        \"INCLUDE_AGENTS\": \"YOUR_AGENTS_LISTING_LOGIC_HERE\"\n      }\n    }\n  }\n}\n```\n\nINCLUDE_AGENTS is one of \"public\", \"private\", \"any\", to list out all public, private, or both (any) agents in the MCP directory.\n\n## API Usage Examples\n\n### Extract Web Text\n\n```javascript\nconst options = {\n  method: 'POST',\n  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},\n  body: '{\"url\":\"https://agent.ai\",\"mode\":\"scrape\"}'\n};\n\nfetch('https://api-lr.agent.ai/v1/action/grab_web_text', options)\n  .then(response => response.json())\n  .then(response => console.log(response))\n  .catch(err => console.error(err));\n```\n\n### Capture Web Screenshot\n\n```javascript\nconst options = {\n  method: 'POST',\n  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},\n  body: '{\"url\":\"https://agent.ai\",\"ttl_for_screenshot\":86400}'\n};\n\nfetch('https://api-lr.agent.ai/v1/action/grab_web_screenshot', options)\n  .then(response => response.json())\n  .then(response => console.log(response))\n  .catch(err => console.error(err));\n```\n\n### Get YouTube Transcript\n\n```javascript\nconst options = {\n  method: 'POST',\n  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},\n  body: '{\"url\":\"https://youtube.com/watch?v=example\"}'\n};\n\nfetch('https://api-lr.agent.ai/v1/action/get_youtube_transcript', options)\n  .then(response => response.json())\n  .then(response => console.log(response))\n  .catch(err => console.error(err));\n```\n\n## Build\n\nDocker build:\n\n```bash\ndocker build -t mcp/agentai:latest .\n```\n\n## License\n\nThis MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, subject to the terms and conditions of the MIT License. For more details, please see the LICENSE file in the project repository.","readmeFilename":"README.md"}