simplehttpskill_skill

Overview

This skill provides a lightweight HTTP client that makes GET, POST, PUT, PATCH, DELETE, HEAD and OPTIONS requests without external dependencies. It supports custom headers, JSON body serialization, automatic retries with exponential backoff and jitter, and always returns an inspectable response object instead of throwing. Use it to call APIs, send webhooks, or fetch URLs reliably from Node.js environments.

How this skill works

Create a client instance with optional default headers and options, then call convenience methods (get, post, put, patch, delete, head, options) or the generic request() method. Requests accept per-call overrides (headers, body, timeout, maxRetries) and bodies that are auto-serialized to JSON when appropriate. The client retries transient failures (network errors, timeouts, HTTP 429 and 5xx) using exponential backoff with jitter and never throws — every call resolves to an object containing ok, status, headers, body, and error.

When to use it

• Calling third-party REST APIs from server-side code without adding dependencies
• Sending webhooks or notifications where retries and backoff are required
• Fetching resources with custom headers or timeouts from Node.js scripts
• Making safe background requests where you prefer an inspectable response over exceptions
• Automating HTTP calls in environments that restrict external libraries

Best practices

• Set defaultHeaders on the client for shared auth headers and override per-request when needed
• Provide JSON bodies as objects to enable automatic serialization and proper content-type handling
• Tune maxRetries and timeout per request for endpoints with different SLAs
• Inspect resp.ok and resp.status to distinguish HTTP errors from network failures
• Avoid retrying non-transient 4xx responses; rely on the client defaults which do not retry them

Example use cases

• POST a webhook payload with retries and backoff to a third-party endpoint
• Fetch paginated API data using GET with a shared Authorization header
• Send periodic heartbeat requests to an internal service and handle transient failures gracefully
• Update a remote resource with PUT or PATCH while attaching request-specific headers
• Perform HEAD or OPTIONS checks for downstream service health and capabilities