SU8 Codes

Blog

Official article

Latest updates

MCP Transports: stdio vs SSE vs Streamable HTTP

MCP transports compared: stdio for local single-user servers, Streamable HTTP for remote or shared deployments, and SSE as a deprecated legacy compatibility path. This guide covers configuration in Claude Code and Cursor, migration steps, and common connection errors. Specification version as of November 15, 2026.

OpenAI API Alternatives: Switching base_url Without Rewriting Code

OpenAI API alternatives fall into two groups: services that implement an OpenAI-compatible endpoint, where you usually change only the base_url, key, and model name, and services with a different API that require code changes. This guide covers documented compatible endpoints, the checks that matter before switching, and common pitfalls. Based on official docs as of November 15, 2026.

Anthropic API Overloaded (529): What It Means and How to Handle It

Anthropic API overloaded (HTTP 529) means the service is temporarily busy, not that your request or key is wrong. This guide explains the official overloaded_error definition, how 529 differs from 429 and other statuses, what the JSON error body and request-id look like, when to retry with exponential backoff and jitter, how the official Python and TypeScript SDKs retry by default, and when to fall back, queue, or open a circuit. Based on Anthropic's official error, rate-limit, streaming, and SDK documentation plus the Python SDK source, checked November 15, 2026.

OpenAI Function Calling Tutorial With a Working Example

OpenAI function calling example with runnable curl and Python code. The tutorial covers tool definitions in JSON Schema, the full request → tool_calls → execute → tool message → final answer loop, multi-turn and parallel calls, argument validation, error returns, and round caps, plus how function calling differs from MCP and plugins. Based on the OpenAI Cookbook, openai-python type definitions, and MCP documentation, checked November 15, 2026.

How to Create Claude Code Custom Slash Commands

A step-by-step guide to Claude Code custom slash commands: project and personal file locations, frontmatter fields, $ARGUMENTS and positional parameters, namespaces, three copyable examples, and fixes for common errors. Checked against the official documentation on November 15, 2026.

Fixing MCP Server Connection Errors in Claude Code

MCP server failed to connect in Claude Code? This troubleshooting guide starts with a symptom/cause/first-check table, then covers config locations and scopes, command paths and runtimes, stdio vs HTTP/SSE/WebSocket transports, project approval and permissions, startup and tool timeouts, debug logs, environment variables and working directories, OAuth and callback ports, and proxy/CA issues, with commands and expected output from the official docs. Based on Claude Code documentation and MCP specification 2026-07-28, checked November 15, 2026.

Claude Code Commands and Slash Commands Cheat Sheet

A Claude Code commands cheat sheet covering launch flags, daily slash commands, keyboard shortcuts, input prefixes, CLAUDE.md and memory, plus hooks and subagents. Every entry is checked against the official documentation as of November 15, 2026, with aliases and removed commands called out.

Codex config.toml Reference

A field-by-field guide to Codex's config.toml: where user and project files live, which layer wins, and the accepted values and minimal examples for model, provider, reasoning effort, approvals, sandboxing, profiles, MCP servers, notify, and history. Checked against OpenAI's documentation and the openai/codex repository on November 15, 2026.

What Is MCP and Why Coding Agents Use It

MCP (Model Context Protocol) is an open standard for connecting AI applications to external tools and data. This guide explains the host/client/server model, the tools/resources/prompts primitives, how coding agents use MCP with databases, docs and APIs, how it differs from function calling and plugins, and what to check before connecting a server. Specification version as of November 15, 2026.

How to Reduce LLM API Cost for Coding Agents

Coding agents can burn money through long context, cache misses, retries, and per-use tools. This guide shows how to read four token prices, tier models, manage context and prompt caching, use batch and off-peak pricing, and estimate a monthly bill. Prices as of November 15, 2026.

OpenAI-Compatible APIs Explained

What an OpenAI-compatible API is, how to connect with just a base_url, an API key and a model name, and where providers differ: base_url paths, feature subsets and model lists. As of November 15, 2026.

Best LLM for Coding: Choose by Task

A task-type guide to choosing among the Claude, GPT, Gemini, DeepSeek, GLM and Kimi lineups, plus effort levels, pricing mechanics and how to read coding benchmarks. As of November 15, 2026.

Codex CLI Tutorial: Install, Configure, Use

Install Codex CLI, sign in, run a first task, set approvals and sandbox in config.toml, learn the daily commands, and add a custom model provider. Install tested on 0.169.0.

How to Install Codex CLI

Install Codex CLI on macOS or Linux with the official installer, or with npm. The installer was tested on macOS and Ubuntu, npm on Ubuntu (0.169.0); Windows and Homebrew weren't tested.