How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

prd - Prompt - OpenSmartRoute

Name: prd
Author: EuclideanAI

Prompt file imported from EuclideanAI/mcp-client (.github/prompts/prd.prompt.md). Copyright stays with the author.

Design and building a custom MCP Client interface that integrates with remote MCP Servers and have LLM agent chat capabilities using Vercel AI SDK. Functional requirements:

Web-based interface for MCP Client
Include Remote MCP Server integration e.g. Jira, Confluence, Github, Slack, Postgres, etc.
LLM agent chat capabilities using Vercel AI SDK
Support for multiple LLMs e.g. Google AI studio, AWS Bedrock Claude
Only need to cover remote MCP server instance, don't need to consider local MCP server instance via Standard I/O connection
Have the ability to connect to remote MCP servers and retrieve and display a list of available tools
Have the ability to pick a tool and execute via the interface with a form entry depending on the required parameters for each MCP tool
LLM agent chat interface should be able to understand the user prompt and pick the right MCP tool to execute
LLM agent should ask for user's permission before executing any MCP tool
LLM agent chat should return the results from the tool execution and summarise where needed for the user Non-functional requirements:
The interface should be responsive
Include dark mode and light mode

Design and building a custom MCP Client interface that integrates with remote MCP Servers and have LLM agent chat capabilities using Vercel AI SDK. Functional requirements: - Web-based interface for MCP Client - Include Remote MCP Server integration e.g. Jira, Confluence, Github, Slack, Postgres, etc. - LLM agent chat capabilities using Vercel AI SDK - Support for multiple LLMs e.g. Google AI studio, AWS Bedrock Claude - Only need to cover remote MCP server instance, don't need to consider local MCP server instance via Standard I/O connection - Have the ability to connect to remote MCP servers and retrieve and display a list of available tools - Have the ability to pick a tool and execute via the interface with a form entry depending on the required parameters for each MCP tool - LLM agent chat interface should be able to understand the user prompt and pick the right MCP tool to execute - LLM agent should ask for user's permission before executing any MCP tool - LLM agent chat should return the results from the tool execution and summarise where needed for the user Non-functional requirements: - The interface should be responsive - Include dark mode and light mode

Install prd and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "prompt": "Design and building a custom MCP Client interface that integrates with remote MCP Servers and have LLM agent chat capabilities using Vercel AI SDK.\nFunctional requirements:\n- Web-based interface for MCP Client\n- Include Remote MCP Server integration e.g. Jira, Confluence, Github, Slack, Postgres, etc.\n- LLM agent chat capabilities using Vercel AI SDK \n- Support for multiple LLMs e.g. Google AI studio, AWS Bedrock Claude\n- Only need to cover remote MCP server instance, don't need to consider local MCP server instance via Standard I/O connection\n- Have the ability to connect to remote MCP servers and retrieve and display a list of available tools\n- Have the ability to pick a tool and execute via the interface with a form entry depending on the required parameters for each MCP tool\n- LLM agent chat interface should be able to understand the user prompt and pick the right MCP tool to execute\n- LLM agent should ask for user's permission before executing any MCP tool\n- LLM agent chat should return the results from the tool execution and summarise where needed for the user\nNon-functional requirements:\n- The interface should be responsive\n- Include dark mode and light mode", "variables": [], "notes": "Design and building a custom MCP Client interface that integrates with remote MCP Servers and have LLM agent chat capabilities using Vercel AI SDK. Functional requirements: - Web-based interface for MCP Client - Include Remote MCP Server integration e.g. Jira, Confluence, Github, Slack, Postgres, etc. - LLM agent chat capabilities using Vercel AI SDK - Support for multiple LLMs e.g. Google AI studio, AWS Bedrock Claude - Only need to cover remote MCP server instance, don't need to consider local MCP server instance via Standard I/O connection - Have the ability to connect to remote MCP servers and retrieve and display a list of available tools - Have the ability to pick a tool and execute via the interface with a form entry depending on the required parameters for each MCP tool - LLM agent chat interface should be able to understand the user prompt and pick the right MCP tool to execute - LLM agent should ask for user's permission before executing any MCP tool - LLM agent chat should return the results from t", "metadata": { "source": { "provider": "github-prompts", "repository": "https://github.com/EuclideanAI/mcp-client", "path": ".github/prompts/prd.prompt.md", "ref": "2ff319668e77f418a41c949e5a0ffd59b9727eef", "url": "https://github.com/EuclideanAI/mcp-client/blob/2ff319668e77f418a41c949e5a0ffd59b9727eef/.github/prompts/prd.prompt.md", "key": "EuclideanAI/mcp-client/.github/prompts/prd.prompt.md" } } }

prd

About

Use it

Install prd and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Manifest

Reviews