Brightsy MCP Server

brightsy

Proxy requests to Brightsy AI agents through the Model Context Protocol, enabling message exchange between users and AI agents for interactive communication.

Instructions

Proxy requests to an Brightsy AI agent

Input Schema

TableJSON Schema

Name	Required	Description	Default
`messages`	Yes	The messages to send to the agent

Implementation Reference

src/index.ts:150-383 (handler)

The main asynchronous handler function for the 'brightsy' tool. It proxies user messages to the Brightsy AI agent API via HTTP POST, maintains conversation history across calls, handles special test commands (echo, history, simulate, clear), processes and formats responses into MCP content format, and includes comprehensive logging and error handling.

async ({ messages }) => {
  try {
    console.error(`Agent proxy tool called with messages:`);
    console.error(JSON.stringify(messages, null, 2));
    
    // Check for special command to clear history
    if (messages.length === 1 && 
        messages[0].role === 'user' && 
        typeof messages[0].content === 'string' && 
        messages[0].content.trim().toLowerCase() === 'clear history') {
      conversationHistory = [];
      console.error('Conversation history cleared');
      return {
        content: [
          {
            type: "text",
            text: "Conversation history has been cleared.",
          },
        ],
      };
    }
    
    // Check for test commands
    if (messages.length === 1 && 
        messages[0].role === 'user' && 
        typeof messages[0].content === 'string') {
      const content = messages[0].content.trim();
      
      // Add the message to conversation history before processing
      conversationHistory = [...conversationHistory, messages[0]];
      
      // Handle test:echo command
      if (content.startsWith('test:echo ')) {
        const message = content.substring('test:echo '.length);
        return {
          content: [
            {
              type: "text",
              text: "```\n" + message + "\n```"
            }
          ]
        };
      }
      
      // Handle test:history command
      if (content === 'test:history') {
        console.error(`Checking history with ${conversationHistory.length} messages:`);
        conversationHistory.forEach((msg, i) => {
          console.error(`[${i}] ${msg.role}: ${typeof msg.content === 'string' ? msg.content : JSON.stringify(msg.content)}`);
        });
        
        if (conversationHistory.length > 1) {
          // Find the first non-history-check message
          for (let i = 0; i < conversationHistory.length - 1; i++) {
            const msg = conversationHistory[i];
            console.error(`Checking message ${i}: ${msg.role} - ${typeof msg.content === 'string' ? msg.content : JSON.stringify(msg.content)}`);
            if (msg.role === 'user' && typeof msg.content === 'string' && !msg.content.includes('test:history')) {
              console.error(`Found message: ${msg.content}`);
              return {
                content: [
                  {
                    type: "text",
                    text: msg.content
                  }
                ]
              };
            }
          }
        }
        console.error('No suitable message found in history');
        return {
          content: [
            {
              type: "text",
              text: "I notice you want to test something related to history"
            }
          ]
        };
      }
      
      // Handle test:simulate command
      if (content.startsWith('test:simulate ')) {
        const message = content.substring('test:simulate '.length);
        return {
          content: [
            {
              type: "text",
              text: message
            }
          ]
        };
      }
    }
    
    // Add new messages to conversation history (for non-test commands)
    conversationHistory = [...conversationHistory, ...messages];
    console.error(`Using conversation history with ${conversationHistory.length} messages`);
    
    const agentUrl = `${agentApiBaseUrl}/api/v1beta/agent/${agent_id}/chat/completion`;
    
    console.error(`Forwarding request to agent: ${agent_id}`);
    console.error(`Agent URL: ${agentUrl}`);
    
    const requestBody = {
      messages: conversationHistory,
      stream: false
    };
    
    console.error(`Request body: ${JSON.stringify(requestBody, null, 2)}`);
    
    const response = await fetch(agentUrl, {
      method: 'POST',
      headers: {
        'Content-Type': 'application/json',
        'Authorization': `Bearer ${api_key}`
      },
      body: JSON.stringify(requestBody)
    });
    
    console.error(`Response status: ${response.status} ${response.statusText}`);
    console.error(`Response headers: ${JSON.stringify(Object.fromEntries([...response.headers]), null, 2)}`);
    
    // Clone the response to log the raw response body
    const responseClone = response.clone();
    const rawResponseText = await responseClone.text();
    console.error(`Raw response body: ${rawResponseText}`);
    
    if (!response.ok) {
      const errorText = await response.text();
      console.error(`Error from agent: ${response.status} ${errorText}`);
      
      return {
        content: [
          {
            type: "text",
            text: `Error from agent: ${errorText}`,
          },
        ],
      };
    }
    
    // Try to parse the response as JSON
    let data;
    try {
      data = JSON.parse(rawResponseText);
      console.error(`Response received from agent and parsed as JSON`);
    } catch (parseError) {
      console.error(`Failed to parse response as JSON: ${parseError}`);
      return {
        content: [
          {
            type: "text",
            text: `Error parsing response: ${parseError}\nRaw response: ${rawResponseText}`,
          },
        ],
      };
    }
    
    console.error(`Response data: ${JSON.stringify(data, null, 2)}`);
    
    // Extract the assistant's response
    const assistantMessage = data.choices?.[0]?.message;
    
    if (!assistantMessage) {
      console.error(`No assistant message found in response: ${JSON.stringify(data, null, 2)}`);
      return {
        content: [
          {
            type: "text",
            text: "No message in agent response",
          },
        ],
      };
    }
    
    console.error(`Assistant message: ${JSON.stringify(assistantMessage, null, 2)}`);
    
    // Add the assistant's response to the conversation history
    if (assistantMessage) {
      conversationHistory.push({
        role: assistantMessage.role || 'assistant',
        content: assistantMessage.content
      });
      console.error(`Added assistant response to history. History now has ${conversationHistory.length} messages`);
    }
    
    // Handle the case where content is already an array of content blocks
    if (Array.isArray(assistantMessage.content)) {
      console.error(`Content is an array, processing directly`);
      // Map the content array to the expected format
      const processedContent = assistantMessage.content.map((item: any) => {
        if (typeof item === 'string') {
          return { type: "text", text: item };
        }
        
        // Handle content blocks (text, image, etc.)
        if (item.text) {
          return { type: "text", text: item.text };
        }
        
        // For other types, convert to string representation
        return { 
          type: "text", 
          text: `[${item.type || 'unknown'} content: ${JSON.stringify(item)}]` 
        };
      });
      
      console.error(`Directly processed content: ${JSON.stringify(processedContent, null, 2)}`);
      
      return {
        content: processedContent,
      };
    }
    
    // Process the content from the assistant's message (for string or other formats)
    const processedContent = processContent(assistantMessage.content);
    console.error(`Processed content: ${JSON.stringify(processedContent, null, 2)}`);
    
    return {
      content: processedContent,
    };
  } catch (error) {
    console.error('Error forwarding request:', error);
    
    return {
      content: [
        {
          type: "text",
          text: `Error: ${error instanceof Error ? error.message : String(error)}`,
        },
      ],
    };
  }
},

src/index.ts:143-148 (schema)

Zod input schema for the tool: an array of messages, each with 'role' (string) and 'content' (string or array).

messages: z.array(
  z.object({
    role: z.string().describe("The role of the message sender"),
    content: z.union([z.string(), z.array(z.any())]).describe("The content of the message")
  })
).describe("The messages to send to the agent")

src/index.ts:139-384 (registration)

MCP tool registration using server.tool(), with dynamic tool name (defaults to 'brightsy'), description 'Proxy requests to an Brightsy AI agent', the input schema, and the handler function.

server.tool(
  tool_name,
  `Proxy requests to an Brightsy AI agent`,
  {
    messages: z.array(
      z.object({
        role: z.string().describe("The role of the message sender"),
        content: z.union([z.string(), z.array(z.any())]).describe("The content of the message")
      })
    ).describe("The messages to send to the agent")
  },
  async ({ messages }) => {
    try {
      console.error(`Agent proxy tool called with messages:`);
      console.error(JSON.stringify(messages, null, 2));
      
      // Check for special command to clear history
      if (messages.length === 1 && 
          messages[0].role === 'user' && 
          typeof messages[0].content === 'string' && 
          messages[0].content.trim().toLowerCase() === 'clear history') {
        conversationHistory = [];
        console.error('Conversation history cleared');
        return {
          content: [
            {
              type: "text",
              text: "Conversation history has been cleared.",
            },
          ],
        };
      }
      
      // Check for test commands
      if (messages.length === 1 && 
          messages[0].role === 'user' && 
          typeof messages[0].content === 'string') {
        const content = messages[0].content.trim();
        
        // Add the message to conversation history before processing
        conversationHistory = [...conversationHistory, messages[0]];
        
        // Handle test:echo command
        if (content.startsWith('test:echo ')) {
          const message = content.substring('test:echo '.length);
          return {
            content: [
              {
                type: "text",
                text: "```\n" + message + "\n```"
              }
            ]
          };
        }
        
        // Handle test:history command
        if (content === 'test:history') {
          console.error(`Checking history with ${conversationHistory.length} messages:`);
          conversationHistory.forEach((msg, i) => {
            console.error(`[${i}] ${msg.role}: ${typeof msg.content === 'string' ? msg.content : JSON.stringify(msg.content)}`);
          });
          
          if (conversationHistory.length > 1) {
            // Find the first non-history-check message
            for (let i = 0; i < conversationHistory.length - 1; i++) {
              const msg = conversationHistory[i];
              console.error(`Checking message ${i}: ${msg.role} - ${typeof msg.content === 'string' ? msg.content : JSON.stringify(msg.content)}`);
              if (msg.role === 'user' && typeof msg.content === 'string' && !msg.content.includes('test:history')) {
                console.error(`Found message: ${msg.content}`);
                return {
                  content: [
                    {
                      type: "text",
                      text: msg.content
                    }
                  ]
                };
              }
            }
          }
          console.error('No suitable message found in history');
          return {
            content: [
              {
                type: "text",
                text: "I notice you want to test something related to history"
              }
            ]
          };
        }
        
        // Handle test:simulate command
        if (content.startsWith('test:simulate ')) {
          const message = content.substring('test:simulate '.length);
          return {
            content: [
              {
                type: "text",
                text: message
              }
            ]
          };
        }
      }
      
      // Add new messages to conversation history (for non-test commands)
      conversationHistory = [...conversationHistory, ...messages];
      console.error(`Using conversation history with ${conversationHistory.length} messages`);
      
      const agentUrl = `${agentApiBaseUrl}/api/v1beta/agent/${agent_id}/chat/completion`;
      
      console.error(`Forwarding request to agent: ${agent_id}`);
      console.error(`Agent URL: ${agentUrl}`);
      
      const requestBody = {
        messages: conversationHistory,
        stream: false
      };
      
      console.error(`Request body: ${JSON.stringify(requestBody, null, 2)}`);
      
      const response = await fetch(agentUrl, {
        method: 'POST',
        headers: {
          'Content-Type': 'application/json',
          'Authorization': `Bearer ${api_key}`
        },
        body: JSON.stringify(requestBody)
      });
      
      console.error(`Response status: ${response.status} ${response.statusText}`);
      console.error(`Response headers: ${JSON.stringify(Object.fromEntries([...response.headers]), null, 2)}`);
      
      // Clone the response to log the raw response body
      const responseClone = response.clone();
      const rawResponseText = await responseClone.text();
      console.error(`Raw response body: ${rawResponseText}`);
      
      if (!response.ok) {
        const errorText = await response.text();
        console.error(`Error from agent: ${response.status} ${errorText}`);
        
        return {
          content: [
            {
              type: "text",
              text: `Error from agent: ${errorText}`,
            },
          ],
        };
      }
      
      // Try to parse the response as JSON
      let data;
      try {
        data = JSON.parse(rawResponseText);
        console.error(`Response received from agent and parsed as JSON`);
      } catch (parseError) {
        console.error(`Failed to parse response as JSON: ${parseError}`);
        return {
          content: [
            {
              type: "text",
              text: `Error parsing response: ${parseError}\nRaw response: ${rawResponseText}`,
            },
          ],
        };
      }
      
      console.error(`Response data: ${JSON.stringify(data, null, 2)}`);
      
      // Extract the assistant's response
      const assistantMessage = data.choices?.[0]?.message;
      
      if (!assistantMessage) {
        console.error(`No assistant message found in response: ${JSON.stringify(data, null, 2)}`);
        return {
          content: [
            {
              type: "text",
              text: "No message in agent response",
            },
          ],
        };
      }
      
      console.error(`Assistant message: ${JSON.stringify(assistantMessage, null, 2)}`);
      
      // Add the assistant's response to the conversation history
      if (assistantMessage) {
        conversationHistory.push({
          role: assistantMessage.role || 'assistant',
          content: assistantMessage.content
        });
        console.error(`Added assistant response to history. History now has ${conversationHistory.length} messages`);
      }
      
      // Handle the case where content is already an array of content blocks
      if (Array.isArray(assistantMessage.content)) {
        console.error(`Content is an array, processing directly`);
        // Map the content array to the expected format
        const processedContent = assistantMessage.content.map((item: any) => {
          if (typeof item === 'string') {
            return { type: "text", text: item };
          }
          
          // Handle content blocks (text, image, etc.)
          if (item.text) {
            return { type: "text", text: item.text };
          }
          
          // For other types, convert to string representation
          return { 
            type: "text", 
            text: `[${item.type || 'unknown'} content: ${JSON.stringify(item)}]` 
          };
        });
        
        console.error(`Directly processed content: ${JSON.stringify(processedContent, null, 2)}`);
        
        return {
          content: processedContent,
        };
      }
      
      // Process the content from the assistant's message (for string or other formats)
      const processedContent = processContent(assistantMessage.content);
      console.error(`Processed content: ${JSON.stringify(processedContent, null, 2)}`);
      
      return {
        content: processedContent,
      };
    } catch (error) {
      console.error('Error forwarding request:', error);
      
      return {
        content: [
          {
            type: "text",
            text: `Error: ${error instanceof Error ? error.message : String(error)}`,
          },
        ],
      };
    }
  },
);

src/index.ts:104-136 (helper)

Helper function processContent() that normalizes agent response content into an array of {type: 'text', text: string} blocks, handling strings, arrays, and content objects.

function processContent(content: any): { type: "text"; text: string }[] {
  if (!content) {
    return [{ type: "text", text: "No content in response" }];
  }
  
  // If content is a string, return it as text
  if (typeof content === 'string') {
    return [{ type: "text", text: content }];
  }
  
  // If content is an array, process each item
  if (Array.isArray(content)) {
    return content.map(item => {
      if (typeof item === 'string') {
        return { type: "text", text: item };
      }
      
      // Handle content blocks (text, image, etc.)
      if (item.text) {
        return { type: "text", text: item.text };
      }
      
      // For other types, convert to string representation
      return { 
        type: "text", 
        text: `[${item.type} content: ${JSON.stringify(item)}]` 
      };
    });
  }
  
  // If we can't process it, return a string representation
  return [{ type: "text", text: JSON.stringify(content) }];
}

Tool Definition Quality

C2.7/5.0

Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries full burden for behavioral disclosure. It mentions 'proxy requests' which implies some form of communication forwarding, but doesn't describe authentication requirements, rate limits, error handling, response format, or what the Brightsy AI agent actually does. This leaves significant behavioral gaps for a proxying tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is extremely concise at just 6 words, with zero wasted language. It's front-loaded with the core purpose and contains no unnecessary elaboration. This is an example of efficient communication that earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a proxying tool with no annotations and no output schema, the description is insufficiently complete. It doesn't explain what the Brightsy AI agent is, what types of requests are proxied, what authentication is needed, or what format the responses take. The combination of vague purpose and missing behavioral context creates significant gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents the single 'messages' parameter with its structure. The description adds no additional parameter semantics beyond what's in the schema. The baseline of 3 is appropriate when the schema does all the parameter documentation work.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose3/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states the tool 'proxy requests to an Brightsy AI agent', which provides a basic verb+resource combination. However, it's vague about what 'proxy requests' specifically entails - whether it's for chat, API calls, or other interactions. Without sibling tools, differentiation isn't needed, but the purpose lacks specificity about the nature of the proxying.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives, nor any context about prerequisites or appropriate scenarios. With no sibling tools, the absence of explicit 'when-not-to-use' guidance is less critical, but there's still no usage context provided.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

brightsyC

Latest Blog Posts

Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
open source
OpenAI
Tool Definition Quality Score (TDQS)
By punkpeye on April 3, 2026.
mcp
The Hackers Who Tracked My Sleep Cycle
By punkpeye on March 26, 2026.
security

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/mattlevine/brightsy-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server