Skip to main content
Glama

shell_grep

Search for text patterns within files using command-line arguments to locate specific content efficiently.

Instructions

Search text patterns in files

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
argsNoCommand arguments

Implementation Reference

  • MCP tool call handler that processes shell_grep by mapping 'shell_grep' to 'shell.grep' config, validating, and executing the 'grep' command via executor.
    server.setRequestHandler(CallToolRequestSchema, async (request) => {
      try {
        const command = String(request.params?.name || '');
        const fullCommand = `shell.${command.replace('shell_', '')}`;  // Replace shell_ back to shell.
        
        if (!(fullCommand in allowedCommands)) {
          return {
            content: [{ type: "text", text: `Unknown command: ${command}` }],
            isError: true
          };
        }
        
        const actualCommand = allowedCommands[fullCommand].command;
        const args = Array.isArray(request.params?.arguments?.args)
          ? request.params.arguments.args.map(String)
          : [];
      
        validator.validateCommand(actualCommand, args);
        const stream = await executor.execute(actualCommand, args);
      
        return {
          content: [{
            type: "text",
            text: await new Promise((resolve, reject) => {
              const chunks: Buffer[] = [];
              stream.stdout.on('data', (chunk: Buffer) => chunks.push(chunk));
              stream.stdout.on('end', () => resolve(Buffer.concat(chunks).toString()));
              stream.stdout.on('error', reject);
            })
          }]
        };
      } catch (error) {
        return {
          content: [{ 
            type: "text", 
            text: `Command execution failed: ${error instanceof Error ? error.message : String(error)}` 
          }],
          isError: true
        };
      }
    });
  • src/index.ts:27-43 (registration)
    Registers all shell tools including 'shell_grep' (from 'shell.grep') in the MCP tools list.
    server.setRequestHandler(ListToolsRequestSchema, async () => {
      const tools = Object.entries(allowedCommands).map(([name, config]) => ({
        name: name.replace('shell.', 'shell_'),  // Replace shell. with shell_
        description: config.description,
        inputSchema: {
          type: "object",
          properties: {
            args: {
              type: "array",
              items: { type: "string" },
              description: "Command arguments"
            }
          }
        }
      }));
      return { tools };
    });
  • Input schema definition for shell_grep tool: expects an object with 'args' array of strings.
    inputSchema: {
      type: "object",
      properties: {
        args: {
          type: "array",
          items: { type: "string" },
          description: "Command arguments"
        }
      }
    }
  • Allowlist configuration for shell.grep (maps to shell_grep tool), defining command, description, allowed arguments, and timeout.
    'shell.grep': {
      command: 'grep',
      description: 'Search text patterns in files',
      allowedArgs: [
        '-i',      // case-insensitive
        '-v',      // invert match
        '-n',      // show line numbers
        '-r',      // recursive
        '-l',      // show file names only
        '--color', // colorize
        '*'        // allow patterns and file paths
      ],
      timeout: 5000
    },
  • Core execution helper that spawns the actual 'grep' process for the shell_grep tool using child_process.spawn, handles streaming output.
    async execute(
      command: string,
      args: string[] = [],
      options: ExecuteOptions = {}
    ): Promise<{ stdout: Readable }> {
      const commandKey = `${command} ${args.join(' ')}`;
      
      try {
        // Check security
        await this.securityChecker.validateCommand(command, args, options);
    
        // Check cache
        const cached = this.cache.get(commandKey);
        if (cached) {
          this.logger.debug('Using cached command result', { command, args });
          return this.createStreamFromCache(cached);
        }
    
        // Remove 'shell.' prefix for execution
        const baseCommand = command.replace('shell.', '');
    
        // Execute command
        this.logger.debug('Starting command execution', { command, args, options });
        const childProcess = spawn(baseCommand, args, {
          stdio: ['ignore', 'pipe', 'pipe'],
          timeout: options.timeout,
          cwd: options.cwd,
          env: {
            ...process.env,
            ...options.env
          },
          signal: options.signal
        });
    
        this.currentProcess = childProcess;
    
        // Error handling
        childProcess.on('error', (error: Error) => {
          this.logger.error('Command execution error', {
            command,
            args,
            error: error.message
          });
          throw new ToolError(
            'PROCESS_ERROR',
            'Command execution error',
            { command, args, error: error.message }
          );
        });
    
        // Timeout handling
        if (options.timeout) {
          setTimeout(() => {
            if (childProcess.exitCode === null) {
              this.logger.warn('Command execution timeout', {
                command,
                args,
                timeout: options.timeout
              });
              childProcess.kill();
              throw new ToolError(
                'TIMEOUT',
                'Command execution timeout',
                { command, args, timeout: options.timeout }
              );
            }
          }, options.timeout);
        }
    
        if (!childProcess.stdout) {
          throw new ToolError(
            'STREAM_ERROR',
            'Unable to get command output stream',
            { command, args }
          );
        }
    
        // Monitor process status
        childProcess.on('exit', (code, signal) => {
          this.logger.debug('Command execution completed', {
            command,
            args,
            exitCode: code,
            signal
          });
        });
    
        return {
          stdout: childProcess.stdout
        };
    
      } catch (error) {
        this.logger.error('Command execution failed', {
          command,
          args,
          error: error instanceof Error ? error.message : String(error)
        });
        
        throw new ToolError(
          'EXECUTION_ERROR',
          'Command execution failed',
          { 
            command, 
            args, 
            error: error instanceof Error ? error.message : String(error)
          }
        );
      }
    }
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 of behavioral disclosure. It states the tool searches text patterns in files, implying a read-only operation, but does not cover critical aspects like error handling (e.g., if files are missing), performance implications (e.g., for large files), or output format. This leaves significant gaps in understanding how the tool behaves beyond its basic purpose.

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 a single phrase, 'Search text patterns in files', which is front-loaded and wastes no words. It efficiently conveys the core purpose without unnecessary elaboration, making it easy for an agent to parse quickly.

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 tool's complexity (a shell command with one parameter), lack of annotations, and no output schema, the description is incomplete. It does not explain what the tool returns (e.g., matched lines, file names), error conditions, or how it interacts with the system. For a tool that could have varied behavior based on arguments, more context is needed to guide effective use.

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?

The input schema has 100% description coverage, with the 'args' parameter documented as 'Command arguments'. The description does not add any meaning beyond this, such as examples of typical arguments (e.g., patterns, file paths) or syntax details. Given the high schema coverage, the baseline score of 3 is appropriate, as the schema handles parameter documentation adequately.

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 'Search text patterns in files' clearly states the tool's function with a specific verb ('search') and resource ('text patterns in files'), avoiding tautology. However, it does not distinguish this tool from potential siblings like 'shell_find' (which might also search files) or clarify that it uses grep-like functionality, leaving some ambiguity in its exact scope.

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 does not mention sibling tools like 'shell_find' or 'shell_cat' that might overlap in functionality, nor does it specify contexts such as searching within file contents versus metadata. Without any usage instructions, the agent must infer when this tool is appropriate.

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

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/kevinwatt/shell-mcp'

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