Skip to main content
Glama

search_files

Search for specific regex patterns within files in a specified directory using a read-only glob filter. Ideal for locating text matches in file systems.

Instructions

Search for a regex pattern within files in a specified directory (read-only).

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
file_patternNoGlob pattern to filter files (e.g., '*.ts'). Defaults to all files ('*').*
pathNoRelative path of the directory to search in..
regexYesThe regex pattern to search for.

Implementation Reference

  • Main handler function that validates args, compiles regex, finds files with glob, searches content in parallel, collects matches with context lines and errors, returns JSON-structured response.
    export const handleSearchFilesFunc = async (
      deps: SearchFilesDependencies,
      args: unknown,
    ): Promise<LocalMcpResponse> => {
      // Updated response type
      const {
        path: relativePath,
        regex: regexString,
        file_pattern: filePattern,
      } = parseAndValidateArgs(args);
    
      const searchRegex = compileSearchRegex(regexString);
      const allResults: SearchResultItem[] = [];
    
      try {
        const filesToSearch = await findFilesToSearch(deps, relativePath, filePattern);
    
        const searchPromises = filesToSearch.map((absoluteFilePath) =>
          searchFileContent({ deps, absoluteFilePath, searchRegex }),
        );
    
        const resultsPerFile = await Promise.all(searchPromises);
        // Flatten results (which now include potential errors)
        for (const fileResults of resultsPerFile) allResults.push(...fileResults);
      } catch (error: unknown) {
        // Errors from findFilesToSearch or Promise.all rejections (should be less likely now)
        if (error instanceof McpError) throw error;
    
        const errorMessage =
          error instanceof Error ? error.message : 'An unknown error occurred during file search.';
        // Error logged via McpError
        // Include a general error if the whole process fails unexpectedly
        allResults.push({ type: 'error', file: 'general', error: errorMessage });
        // Don't throw, return the collected results including the general error
        // throw new McpError(ErrorCode.InternalError, errorMessage);
      }
    
      // Return the structured data including matches and errors
      return {
        content: [
          {
            type: 'text',
            text: JSON.stringify({ results: allResults }, undefined, 2),
          },
        ],
        data: {
          results: allResults,
        },
      };
    };
  • Zod input schema defining parameters: path (directory), regex (pattern), file_pattern (glob filter).
    export const SearchFilesArgsSchema = z
      .object({
        path: z
          .string()
          .optional()
          .default('.')
          .describe('Relative path of the directory to search in.'),
        regex: z
          .string()
          .min(1, { message: 'Regex pattern cannot be empty' })
          .describe('The regex pattern to search for.'),
        file_pattern: z
          .string()
          .optional()
          .default('*')
          .describe("Glob pattern to filter files (e.g., '*.ts'). Defaults to all files ('*')."),
      })
      .strict();
  • Tool definition object registering 'search_files' with name, description, input/output schemas, and handler wrapper that provides dependencies.
    export const searchFilesToolDefinition = {
      name: 'search_files',
      description:
        'Search for a regex pattern within files in a specified directory (read-only). Returns matches and any errors encountered.',
      inputSchema: SearchFilesArgsSchema,
      // Define output schema
      outputSchema: z.object({
        results: z.array(
          z.object({
            type: z.enum(['match', 'error']),
            file: z.string(),
            line: z.number().int().optional(),
            match: z.string().optional(),
            context: z.array(z.string()).optional(),
            error: z.string().optional(),
          }),
        ),
      }),
      // Use the imported local McpResponse type
      handler: (args: unknown): Promise<LocalMcpResponse> => {
        const deps: SearchFilesDependencies = {
          readFile: async (_path, _options) => {
            const encoding = typeof _options === 'string' ? _options : (_options?.encoding ?? 'utf8');
            return fsPromises.readFile(_path, { encoding });
          },
          glob: globFn,
          resolvePath: resolvePathUtil,
          PROJECT_ROOT: projectRootUtil,
          pathRelative: path.relative.bind(path),
          pathJoin: path.join.bind(path),
        };
        return handleSearchFilesFunc(deps, args);
      },
    };
  • Central registry aggregating all tool definitions, including searchFilesToolDefinition (line 59 in original file).
    export const allToolDefinitions: HandlerToolDefinition[] = [
      listFilesToolDefinition,
      statItemsToolDefinition,
      readContentToolDefinition,
      writeContentToolDefinition,
      deleteItemsToolDefinition,
      createDirectoriesToolDefinition,
      chmodItemsToolDefinition,
      chownItemsToolDefinition,
      moveItemsToolDefinition,
      copyItemsToolDefinition,
      searchFilesToolDefinition,
      replaceContentToolDefinition,
      {
        name: 'apply_diff',
        description: 'Apply diffs to files',
        inputSchema: applyDiffInputSchema,
        handler: async (args: unknown): Promise<McpToolResponse> => {
          const validatedArgs = applyDiffInputSchema.parse(args);
          const result: ApplyDiffOutput = await handleApplyDiff(validatedArgs.changes, {
            readFile: async (path: string) => fs.promises.readFile(path, 'utf8'),
            writeFile: async (path: string, content: string) =>
              fs.promises.writeFile(path, content, 'utf8'),
            path,
            projectRoot: process.cwd(),
          });
          return {
            content: [
              {
                type: 'text',
                text: JSON.stringify(
                  {
                    success: result.success,
                    results: result.results,
                  },
                  undefined,
                  2,
                ),
              },
            ],
          };
        },
      },
    ];
  • Import of the search_files tool definition for inclusion in the central registry.
    import { searchFilesToolDefinition } from './search-files.js';
Behavior3/5

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

With no annotations provided, the description carries the full burden of behavioral disclosure. It adds the '(read-only)' qualifier, which is valuable context about safety. However, it doesn't describe other important behaviors like whether the search is recursive, case-sensitive, or what the output format looks like (matches, line numbers, etc.).

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 a single, efficient sentence with zero wasted words. It's appropriately sized and front-loaded with the core functionality. Every element ('search for regex pattern', 'within files', 'in specified directory', 'read-only') earns its place.

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

Completeness3/5

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

Given the tool's moderate complexity (regex search across files), no annotations, and no output schema, the description is minimally adequate. It covers the basic operation and safety qualifier but lacks details about search behavior (recursive?), output format, or error handling. For a search tool with no structured output documentation, more completeness would be helpful.

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 all three parameters thoroughly. The description adds no additional parameter semantics beyond what's in the schema - it mentions 'regex pattern' and 'specified directory' but provides no extra syntax, format, or usage details. 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.

Purpose4/5

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

The description clearly states the tool's purpose with specific verbs ('search for a regex pattern within files') and resource ('in a specified directory'), and includes the '(read-only)' qualifier. However, it doesn't explicitly differentiate from sibling tools like 'list_files' or 'read_content' beyond the regex search aspect.

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 like 'list_files' for directory listing or 'read_content' for file reading. It mentions the directory scope but offers no explicit when/when-not instructions or sibling tool comparisons.

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/SylphxAI/filesystem-mcp'

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