Skip to main content
Glama
signal-slot

MCP GDB Server

by signal-slot

gdb_backtrace

Analyze and display the call stack during a GDB debugging session, optionally including variables and limiting the number of frames. Simplify debugging by identifying execution flow and context.

Instructions

Show call stack

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
fullNoShow variables in each frame (optional)
limitNoMaximum number of frames to show (optional)
sessionIdYesGDB session ID

Implementation Reference

  • The main handler function for the gdb_backtrace tool. It checks for an active GDB session, builds the backtrace command (with optional full and limit), executes it via executeGdbCommand, and returns the output or error.
    private async handleGdbBacktrace(args: any) {
      const { sessionId, full = false, limit } = args;
      
      if (!activeSessions.has(sessionId)) {
        return {
          content: [
            {
              type: 'text',
              text: `No active GDB session with ID: ${sessionId}`
            }
          ],
          isError: true
        };
      }
      
      const session = activeSessions.get(sessionId)!;
      
      try {
        // Build backtrace command with options
        let command = full ? "backtrace full" : "backtrace";
        if (typeof limit === 'number') {
          command += ` ${limit}`;
        }
        
        const output = await this.executeGdbCommand(session, command);
        
        return {
          content: [
            {
              type: 'text',
              text: `Backtrace${full ? ' (full)' : ''}${limit ? ` (limit: ${limit})` : ''}:\n\n${output}`
            }
          ]
        };
      } catch (error) {
        const errorMessage = error instanceof Error ? error.message : String(error);
        return {
          content: [
            {
              type: 'text',
              text: `Failed to get backtrace: ${errorMessage}`
            }
          ],
          isError: true
        };
      }
    }
  • The tool registration object including name, description, and input schema definition for gdb_backtrace.
      name: 'gdb_backtrace',
      description: 'Show call stack',
      inputSchema: {
        type: 'object',
        properties: {
          sessionId: {
            type: 'string',
            description: 'GDB session ID'
          },
          full: {
            type: 'boolean',
            description: 'Show variables in each frame (optional)'
          },
          limit: {
            type: 'number',
            description: 'Maximum number of frames to show (optional)'
          }
        },
        required: ['sessionId']
      }
    },
  • src/index.ts:383-384 (registration)
    The switch case in the request handler that routes 'gdb_backtrace' tool calls to the handleGdbBacktrace method.
    case 'gdb_backtrace':
      return await this.handleGdbBacktrace(request.params.arguments);
  • Helper function used by gdb_backtrace to execute the backtrace GDB command and capture its output with timeout and error handling.
    private executeGdbCommand(session: GdbSession, command: string): Promise<string> {
      return new Promise<string>((resolve, reject) => {
        if (!session.ready) {
          reject(new Error('GDB session is not ready'));
          return;
        }
        
        // Write command to GDB's stdin
        if (session.process.stdin) {
          session.process.stdin.write(command + '\n');
        } else {
          reject(new Error('GDB stdin is not available'));
          return;
        }
        
        let output = '';
        let responseComplete = false;
        
        // Create a one-time event handler for GDB output
        const onLine = (line: string) => {
          output += line + '\n';
          
          // Check if this line indicates the end of the GDB response
          if (line.includes('(gdb)') || line.includes('^done') || line.includes('^error')) {
            responseComplete = true;
            
            // If we've received the complete response, resolve the promise
            if (responseComplete) {
              // Remove the listener to avoid memory leaks
              session.rl.removeListener('line', onLine);
              resolve(output);
            }
          }
        };
        
        // Add the line handler to the readline interface
        session.rl.on('line', onLine);
        
        // Set a timeout to prevent hanging
        const timeout = setTimeout(() => {
          session.rl.removeListener('line', onLine);
          reject(new Error('GDB command timed out'));
        }, 10000); // 10 second timeout
        
        // Handle GDB errors
        const errorHandler = (data: Buffer) => {
          const errorText = data.toString();
          output += `[stderr] ${errorText}\n`;
        };
        
        // Add error handler
        if (session.process.stderr) {
          session.process.stderr.once('data', errorHandler);
        }
        
        // Clean up event handlers when the timeout expires
        timeout.unref();
      });
    }
Behavior2/5

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

No annotations are provided, so the description carries the full burden. 'Show call stack' implies a read-only operation, but it doesn't disclose behavioral traits such as whether it requires an active debug session, if it's safe to call repeatedly, what happens if the session is invalid, or the format of the output. This leaves significant gaps for a debugging 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 with just two words, 'Show call stack', which is front-loaded and wastes no space. Every word earns its place by directly stating the tool's purpose without unnecessary elaboration.

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?

Given the complexity of debugging tools, no annotations, and no output schema, the description is incomplete. It doesn't cover essential context like what the output looks like (e.g., stack trace format), error conditions, or dependencies on other tools (e.g., requiring gdb_attach first). This makes it inadequate for effective use by an AI agent.

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 fully documents the three parameters (full, limit, sessionId). The description adds no additional meaning beyond what the schema provides, such as explaining how 'full' affects variable display or typical values for 'limit'. Baseline 3 is appropriate when the schema does the heavy lifting.

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 'Show call stack' clearly indicates the tool displays the call stack, which is a specific function. However, it doesn't distinguish this from sibling tools like gdb_info_registers or gdb_print that might also display debugging information, nor does it specify the resource (e.g., for a GDB session) beyond what's implied by the tool name.

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. It doesn't mention prerequisites (e.g., needing an active GDB session), context (e.g., during debugging after a breakpoint), or comparisons to siblings like gdb_backtrace vs. gdb_info_registers for different debugging info.

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

Related Tools

Latest Blog Posts

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/signal-slot/mcp-gdb'

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