Skip to main content
Glama
DeanWard

HAL (HTTP API Layer)

HTTP POST Request

http-post

Send HTTP POST requests to specified URLs with optional body, headers, and secret substitution using {secrets.key} syntax, simplifying API interactions within the HAL server environment.

Instructions

Make an HTTP POST request to a specified URL with optional body and headers. Supports secret substitution using {secrets.key} syntax in URL, headers, and body where 'key' corresponds to HAL_SECRET_KEY environment variables.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
bodyNo
contentTypeNoapplication/json
headersNo
urlYes

Implementation Reference

  • src/index.ts:586-605 (registration)
    Registration of the 'http-post' tool using server.registerTool, including title, description, inputSchema, and the handler function.
    server.registerTool(
      "http-post",
      {
        title: "HTTP POST Request",
        description: "Make an HTTP POST request to a specified URL with optional body and headers. Supports secret substitution using {secrets.key} syntax in URL, headers, and body where 'key' corresponds to HAL_SECRET_KEY environment variables.",
        inputSchema: {
          url: z.string().url(),
          body: z.string().optional(),
          headers: z.record(z.string()).optional(),
          contentType: z.string().default('application/json')
        }
      },
      async ({ url, body, headers = {}, contentType }: { url: string; body?: string; headers?: Record<string, string>; contentType: string }) => {
        const requestHeaders = {
          'Content-Type': contentType,
          ...headers
        };
        return makeHttpRequest('POST', url, { headers: requestHeaders, body });
      }
    );
  • Zod input schema for the http-post tool, validating url (required URL), body (optional string), headers (optional record), contentType (default 'application/json').
    inputSchema: {
      url: z.string().url(),
      body: z.string().optional(),
      headers: z.record(z.string()).optional(),
      contentType: z.string().default('application/json')
    }
  • Specific handler for http-post tool: constructs request headers with Content-Type and calls the core makeHttpRequest with method 'POST'.
    async ({ url, body, headers = {}, contentType }: { url: string; body?: string; headers?: Record<string, string>; contentType: string }) => {
      const requestHeaders = {
        'Content-Type': contentType,
        ...headers
      };
      return makeHttpRequest('POST', url, { headers: requestHeaders, body });
    }
  • Core helper function implementing HTTP requests for all http-* tools: handles secret substitution in URL/headers/body/query, URL allow/block lists, performs fetch, processes response (JSON/text), redacts secrets from output.
    async function makeHttpRequest(
      method: string,
      url: string,
      options: {
        headers?: Record<string, string>;
        body?: string;
        queryParams?: Record<string, any>;
      } = {}
    ) {
      try {
        const { headers = {}, body, queryParams = {} } = options;
        
        // First, substitute secrets in URL to get the final URL for validation
        // We need to do this in two passes to handle URL restrictions properly
        const processedUrl = substituteSecrets(url, url);
        
        // Now substitute secrets in headers, body, and query parameters using the processed URL
        const processedHeaders = substituteSecretsInObject(headers, processedUrl);
        const processedBody = body ? substituteSecrets(body, processedUrl) : body;
        const processedQueryParams = substituteSecretsInObject(queryParams, processedUrl);
        
        // Build URL with query parameters
        const urlObj = new URL(processedUrl);
        Object.entries(processedQueryParams).forEach(([key, value]) => {
          if (value !== undefined && value !== null) {
            urlObj.searchParams.set(key, String(value));
          }
        });
        
        const finalUrl = urlObj.toString();
        
        // Check global URL whitelist/blacklist
        const urlCheck = isUrlAllowedGlobal(finalUrl);
        if (!urlCheck.allowed) {
          throw new Error(urlCheck.reason || 'URL is not allowed');
        }
        
        const defaultHeaders = {
          'User-Agent': 'HAL-MCP/1.0.0',
          ...processedHeaders
        };
        
             // Add Content-Type for methods that typically send data
         if (['POST', 'PUT', 'PATCH'].includes(method.toUpperCase()) && processedBody && !('Content-Type' in processedHeaders)) {
           (defaultHeaders as any)['Content-Type'] = 'application/json';
         }
        
        const response = await fetch(finalUrl, {
          method: method.toUpperCase(),
          headers: defaultHeaders,
          body: processedBody
        });
    
        const contentType = response.headers.get('content-type') || 'text/plain';
        let content: string;
        
        // HEAD requests don't have a body by design
        if (method.toUpperCase() === 'HEAD') {
          content = '(No body - HEAD request)';
        } else {
          try {
            if (contentType.includes('application/json')) {
              const text = await response.text();
              if (text.trim()) {
                content = JSON.stringify(JSON.parse(text), null, 2);
              } else {
                content = '(Empty response)';
              }
            } else {
              content = await response.text();
            }
          } catch (parseError) {
            // If JSON parsing fails, try to get text
            try {
              content = await response.text();
            } catch (textError) {
              content = '(Unable to parse response)';
            }
          }
        }
    
        // Redact secrets from response headers and content before returning
        const redactedHeaders = Array.from(response.headers.entries())
          .map(([key, value]) => `${key}: ${redactSecretsFromText(value)}`)
          .join('\n');
        const redactedContent = redactSecretsFromText(content);
    
             return {
           content: [{
             type: "text" as const,
             text: `Status: ${response.status} ${response.statusText}\n\nHeaders:\n${redactedHeaders}\n\nBody:\n${redactedContent}`
           }]
         };
      } catch (error) {
        const errorMessage = error instanceof Error ? error.message : 'Unknown error';
        const redactedErrorMessage = redactSecretsFromText(errorMessage);
             return {
           content: [{
             type: "text" as const,
             text: `Error making ${method.toUpperCase()} request: ${redactedErrorMessage}`
           }],
           isError: true
         };
      }
    }
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. It discloses key behavioral traits: it's a network request tool (implying potential side effects), supports secret substitution, and mentions environment variables. However, it lacks details on error handling, timeouts, authentication needs beyond secrets, rate limits, or response format, which are important for a network 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 appropriately sized and front-loaded, with the core purpose stated first and additional details (secret substitution) added concisely. Every sentence earns its place by providing essential information without redundancy.

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 complexity (network request with 4 parameters, no annotations, no output schema), the description is incomplete. It covers the basic operation and secret feature but omits critical details like expected response format, error handling, and authentication requirements. This leaves gaps for an AI agent to use the tool effectively.

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 0%, so the description must compensate. It adds meaning by explaining the purpose of parameters (e.g., URL for the request, optional body and headers, secret substitution syntax), but does not detail parameter constraints like URL format or header structure. The baseline is lower due to low coverage, and while it provides some context, it doesn't fully document all 4 parameters.

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

Purpose5/5

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

The description clearly states the specific action ('Make an HTTP POST request') with the resource ('to a specified URL'), and distinguishes it from siblings by specifying the HTTP method (POST) rather than GET, DELETE, etc. It goes beyond just restating the name/title by including details about optional body/headers and secret substitution.

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

Usage Guidelines4/5

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

The description provides clear context for when to use this tool (for HTTP POST requests), but does not explicitly state when not to use it or name alternatives. It implies usage for POST operations but lacks explicit guidance on choosing between POST and other HTTP methods like PUT or PATCH from the sibling list.

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/DeanWard/HAL'

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