Skip to main content
Glama
MoralisWeb3

Moralis MCP Server

Official
by MoralisWeb3

evm_getnftmetadata

Fetch comprehensive NFT metadata including on-chain and off-chain data, floor prices, and rarity information for specific tokens across multiple blockchain networks.

Instructions

Fetch metadata for a specific NFT. Includes on-chain metadata as well as off-chain metadata, floor prices, rarity and more where available.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
chainNoThe chain to queryeth
addressYesThe address of the NFT contract
token_idYesThe ID of the token
formatNoThe format of the token IDdecimal
normalizeMetadataNoShould normalized metadata be returned?
media_itemsNoShould preview media data be returned?
include_pricesNoShould NFT last sale prices be included in the result?

Implementation Reference

  • The core handler function that implements all dynamic 'evm_' prefixed tools, including 'evm_getnftmetadata'. It validates inputs, constructs HTTP requests to the Moralis API based on the tool's OpenAPI definition, adds authentication, executes the request via axios, and formats the response.
    export async function executeApiTool(
      toolName: string,
      definition: McpToolDefinition,
      toolArgs: JsonObject,
      allSecuritySchemes: Record<string, any>,
      token?: string,
    ): Promise<CallToolResult> {
      try {
        // Validate arguments against the input schema
        let validatedArgs: JsonObject;
        try {
          const zodSchema = getZodSchemaFromJsonSchema(
            definition.inputSchema,
            toolName,
          );
          const argsToParse =
            typeof toolArgs === 'object' && toolArgs !== null ? toolArgs : {};
          validatedArgs = zodSchema.parse(argsToParse);
        } catch (error: unknown) {
          if (error instanceof ZodError) {
            const validationErrorMessage = `Invalid arguments for tool '${toolName}': ${error.errors.map((e) => `${e.path.join('.')} (${e.code}): ${e.message}`).join(', ')}`;
            return { content: [{ type: 'text', text: validationErrorMessage }] };
          } else {
            const errorMessage =
              error instanceof Error ? error.message : String(error);
            return {
              content: [
                {
                  type: 'text',
                  text: `Internal error during validation setup: ${errorMessage}`,
                },
              ],
            };
          }
        }
    
        // Prepare URL, query parameters, headers, and request body
        let urlPath = definition.pathTemplate;
        const queryParams: Record<string, any> = {};
        const headers: Record<string, string> = {
          Accept: 'application/json',
          'X-Moralis-Platform': 'MCP',
        };
        let requestBodyData: any = undefined;
    
        // Apply parameters to the URL path, query, or headers
        definition.executionParameters.forEach((param) => {
          const value = validatedArgs[param.name];
          if (typeof value !== 'undefined' && value !== null) {
            if (param.in === 'path') {
              urlPath = urlPath.replace(
                `{${param.name}}`,
                encodeURIComponent(String(value)),
              );
            } else if (param.in === 'query') {
              queryParams[param.name] = value;
            } else if (param.in === 'header') {
              headers[param.name.toLowerCase()] = String(value);
            }
          }
        });
    
        // Ensure all path parameters are resolved
        if (urlPath.includes('{')) {
          throw new Error(`Failed to resolve path parameters: ${urlPath}`);
        }
    
        // Construct the full URL
        const requestUrl = `${definition.baseUrl}${urlPath}`;
    
        // Handle request body if needed
        if (
          definition.requestBodyContentType &&
          typeof validatedArgs['requestBody'] !== 'undefined'
        ) {
          requestBodyData = validatedArgs['requestBody'];
          headers['content-type'] = definition.requestBodyContentType;
        }
    
        // Apply security requirements if available
        // Security requirements use OR between array items and AND within each object
        const appliedSecurity = definition.securityRequirements?.find((req) => {
          // Try each security requirement (combined with OR)
          return Object.entries(req).every(([schemeName, scopesArray]) => {
            const scheme = allSecuritySchemes[schemeName];
            if (!scheme) return false;
    
            // API Key security (header, query, cookie)
            if (scheme.type === 'apiKey') {
              return !!token || !!Config.MORALIS_API_KEY;
            }
    
            return false;
          });
        });
    
        // If we found matching security scheme(s), apply them
        if (appliedSecurity) {
          // Apply each security scheme from this requirement (combined with AND)
          for (const [schemeName, scopesArray] of Object.entries(appliedSecurity)) {
            const scheme = allSecuritySchemes[schemeName];
    
            // API Key security
            if (scheme?.type === 'apiKey') {
              const apiKey = token || Config.MORALIS_API_KEY;
              if (apiKey) {
                if (scheme.in === 'header') {
                  headers[scheme.name.toLowerCase()] = apiKey;
                  console.error(
                    `Applied API key '${schemeName}' in header '${scheme.name}'`,
                  );
                }
              }
            }
          }
        }
        // Log warning if security is required but not available
        else if (definition.securityRequirements?.length > 0) {
          // First generate a more readable representation of the security requirements
          const securityRequirementsString = definition.securityRequirements
            .map((req) => {
              const parts = Object.entries(req)
                .map(([name, scopesArray]) => {
                  const scopes = scopesArray as string[];
                  if (scopes.length === 0) return name;
                  return `${name} (scopes: ${scopes.join(', ')})`;
                })
                .join(' AND ');
              return `[${parts}]`;
            })
            .join(' OR ');
    
          console.warn(
            `Tool '${toolName}' requires security: ${securityRequirementsString}, but no suitable credentials found.`,
          );
        }
    
        // Prepare the axios request configuration
        const config: AxiosRequestConfig = {
          method: definition.method.toUpperCase(),
          url: requestUrl,
          params: queryParams,
          headers: headers,
          ...(requestBodyData !== undefined && { data: requestBodyData }),
        };
    
        // Log request info to stderr (doesn't affect MCP output)
        console.error(
          `Executing tool "${toolName}": ${config.method} ${config.url}`,
        );
    
        // Execute the request
        const response = await axios(config);
    
        // Process and format the response
        let responseText = '';
        const contentType = response.headers['content-type']?.toLowerCase() || '';
    
        // Handle JSON responses
        if (
          contentType.includes('application/json') &&
          typeof response.data === 'object' &&
          response.data !== null
        ) {
          try {
            responseText = JSON.stringify(response.data, null, 2);
          } catch (e) {
            responseText = '[Stringify Error]';
          }
        }
        // Handle string responses
        else if (typeof response.data === 'string') {
          responseText = response.data;
        }
        // Handle other response types
        else if (response.data !== undefined && response.data !== null) {
          responseText = String(response.data);
        }
        // Handle empty responses
        else {
          responseText = `(Status: ${response.status} - No body content)`;
        }
    
        // Return formatted response
        return {
          content: [
            {
              type: 'text',
              text: `API Response (Status: ${response.status}):\n${responseText}`,
            },
          ],
        };
      } catch (error: unknown) {
        // Handle errors during execution
        let errorMessage: string;
    
        // Format Axios errors specially
        if (axios.isAxiosError(error)) {
          errorMessage = formatApiError(error);
        }
        // Handle standard errors
        else if (error instanceof Error) {
          errorMessage = error.message;
        }
        // Handle unexpected error types
        else {
          errorMessage = `Unexpected error: ${String(error)}`;
        }
    
        // Log error to stderr
        console.error(
          `Error during execution of tool '${toolName}':`,
          errorMessage,
        );
    
        // Return error message to client
        return { content: [{ type: 'text', text: errorMessage }] };
      }
    }
  • src/server.ts:96-120 (registration)
    Registers the MCP CallToolRequest handler which looks up the dynamic tool definition by name (e.g., 'evm_getnftmetadata') and dispatches to the executeApiTool handler.
    server.setRequestHandler(
      CallToolRequestSchema,
      async (request: CallToolRequest, c): Promise<CallToolResult> => {
        const { name: toolName, arguments: toolArgs } = request.params;
        const toolDefinition = toolDefinitionMap[toolName];
        if (!toolDefinition) {
          console.error(`Error: Unknown tool requested: ${toolName}`);
          return {
            content: [
              {
                type: 'text',
                text: `Error: Unknown tool requested: ${toolName}`,
              },
            ],
          };
        }
        return executeApiTool(
          toolName,
          toolDefinition,
          toolArgs ?? {},
          securitySchemes,
          c.authInfo?.token,
        );
      },
    );
  • src/server.ts:85-94 (registration)
    Registers the ListToolsRequest handler that returns all dynamically extracted tools, including 'evm_getnftmetadata', with their schemas.
    server.setRequestHandler(ListToolsRequestSchema, async () => {
      const toolsForClient: Tool[] = Object.values(toolDefinitionMap).map(
        (def) => ({
          name: def.name,
          description: def.description,
          inputSchema: def.inputSchema,
        }),
      );
      return { tools: toolsForClient };
    });
  • Generates the inputSchema JSON Schema for each tool from the OpenAPI operation's parameters and requestBody, used for 'evm_getnftmetadata' schema.
    export function generateInputSchemaAndDetails(
      operation: OpenAPIV3.OperationObject,
    ): {
      inputSchema: JSONSchema7 | boolean;
      parameters: OpenAPIV3.ParameterObject[];
      requestBodyContentType?: string;
    } {
      const properties: { [key: string]: JSONSchema7 | boolean } = {};
      const required: string[] = [];
    
      // Process parameters
      const allParameters: OpenAPIV3.ParameterObject[] = Array.isArray(
        operation.parameters,
      )
        ? operation.parameters.map((p) => p as OpenAPIV3.ParameterObject)
        : [];
    
      allParameters.forEach((param) => {
        if (!param.name || !param.schema) return;
    
        const paramSchema = mapOpenApiSchemaToJsonSchema(
          param.schema as OpenAPIV3.SchemaObject,
        );
        if (typeof paramSchema === 'object') {
          paramSchema.description = param.description || paramSchema.description;
        }
    
        properties[param.name] = paramSchema;
        if (param.required) required.push(param.name);
      });
    
      // Process request body (if present)
      let requestBodyContentType: string | undefined = undefined;
    
      if (operation.requestBody) {
        const opRequestBody = operation.requestBody as OpenAPIV3.RequestBodyObject;
        const jsonContent = opRequestBody.content?.['application/json'];
        const firstContent = opRequestBody.content
          ? Object.entries(opRequestBody.content)[0]
          : undefined;
    
        if (jsonContent?.schema) {
          requestBodyContentType = 'application/json';
          const bodySchema = mapOpenApiSchemaToJsonSchema(
            jsonContent.schema as OpenAPIV3.SchemaObject,
          );
    
          if (typeof bodySchema === 'object') {
            bodySchema.description =
              opRequestBody.description ||
              bodySchema.description ||
              'The JSON request body.';
          }
    
          properties['requestBody'] = bodySchema;
          if (opRequestBody.required) required.push('requestBody');
        } else if (firstContent) {
          const [contentType] = firstContent;
          requestBodyContentType = contentType;
    
          properties['requestBody'] = {
            type: 'string',
            description:
              opRequestBody.description ||
              `Request body (content type: ${contentType})`,
          };
    
          if (opRequestBody.required) required.push('requestBody');
        }
      }
    
      // Combine everything into a JSON Schema
      const inputSchema: JSONSchema7 = {
        type: 'object',
        properties,
        ...(required.length > 0 && { required }),
      };
    
      return { inputSchema, parameters: allParameters, requestBodyContentType };
    }
  • Extracts all tool definitions from OpenAPI spec, adding 'evm_' prefix to operationIds to create tools like 'evm_getnftmetadata', including method, path, parameters, security.
    export function extractToolsFromApi(
      api: OpenAPIV3DocumentX,
      prefix?: string
    ): McpToolDefinition[] {
      const tools: McpToolDefinition[] = [];
      const usedNames = new Set<string>();
      const globalSecurity = api.security || [];
    
      if (!api.paths) return tools;
    
      for (const [path, pathItem] of Object.entries(api.paths)) {
        if (!pathItem) continue;
    
        for (const method of Object.values(OpenAPIV3.HttpMethods)) {
          const operation = pathItem[method];
          if (!operation) continue;
    
          // Generate a unique name for the tool
          let baseName = operation.operationId || generateOperationId(method, path);
          if (!baseName) continue;
    
          // Sanitize the name to be MCP-compatible (only a-z, 0-9, _, -)
          baseName = baseName
            .replace(/\./g, '_')
            .replace(/[^a-z0-9_-]/gi, '_')
            .toLowerCase();
    
          let finalToolName = baseName;
          let counter = 1;
          while (usedNames.has(finalToolName)) {
            finalToolName = `${baseName}_${counter++}`;
          }
          usedNames.add(finalToolName);
    
          // Get or create a description
          const description =
            operation.description ||
            operation.summary ||
            `Executes ${method.toUpperCase()} ${path}`;
    
          const prompt = operation['x-mcp-prompt'];
    
          // Generate input schema and extract parameters
          const { inputSchema, parameters, requestBodyContentType } =
            generateInputSchemaAndDetails(operation);
    
          // Extract parameter details for execution
          const executionParameters = parameters.map((p) => ({
            name: p.name,
            in: p.in,
          }));
    
          // Determine security requirements
          const securityRequirements =
            operation.security === null
              ? globalSecurity
              : operation.security || globalSecurity;
    
          // Create the tool definition
          tools.push({
            name: prefix + finalToolName,
            description,
            inputSchema,
            method,
            pathTemplate: path,
            parameters,
            executionParameters,
            requestBodyContentType,
            securityRequirements,
            operationId: baseName,
            prompt,
          });
        }
      }
    
      return tools;
    }
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 what data is included (metadata, floor prices, rarity) but doesn't cover important aspects like rate limits, authentication requirements, error handling, response format, or whether this is a read-only operation (though 'fetch' implies read-only). The phrase 'where available' hints at partial responses but isn't specific.

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

Conciseness4/5

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

The description is a single, efficient sentence that communicates the core purpose and scope. It's appropriately front-loaded with the main action. However, it could be slightly more structured by separating primary function from additional features.

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 tool with 7 parameters, no annotations, and no output schema, the description is insufficient. It doesn't explain the response structure, error cases, rate limits, or authentication requirements. While it mentions what data is fetched, it doesn't provide enough context for reliable agent usage given the complexity.

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 all parameters are documented in the schema. The description doesn't add any parameter-specific information beyond what's in the schema (like explaining relationships between parameters or providing examples). The baseline of 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 verb ('fetch') and resource ('metadata for a specific NFT'), and specifies what metadata is included (on-chain, off-chain, floor prices, rarity). It distinguishes from some siblings like 'evm_getnftcollectionstats' by focusing on individual NFTs rather than collections, but doesn't explicitly differentiate from 'evm_getmultiplenfts' or 'solana_getnftmetadata'.

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?

No guidance is provided about when to use this tool versus alternatives like 'evm_getmultiplenfts' for batch operations, 'evm_getnftcollectionstats' for collection-level data, or 'solana_getnftmetadata' for Solana NFTs. The description doesn't mention prerequisites, error conditions, or typical use cases.

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/MoralisWeb3/moralis-mcp-server'

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