LING71671/Open-ClaudeCode

Agent Development explains how to create Claude Code agents, including frontmatter, trigger descriptions, tools, colors, and system prompt structure. Use it when building or refining autonomous agents and subagents for plugins.

Claude Opus 4.5 Migration updates prompts and code from Sonnet 4.0, Sonnet 4.5, or Opus 4.1 to Opus 4.5. It changes model strings, removes unsupported beta headers, and can adjust prompts for known behavior differences.

Command Development shows how to create and organize Claude Code slash commands with Markdown, YAML frontmatter, arguments, and tool settings. Use it when defining reusable commands or adding interactive behavior and file references.

Document Release keeps README files, examples, settings, and related docs aligned with code changes. Use it after behavior, CLI flags, provider support, or plugin surfaces change.

Frontend Design guides the creation of distinctive, production-grade interfaces for components, pages, and applications. Use it when building web UIs that need a clear aesthetic direction and real working code.

GStack QA Only produces read-only QA reports for an OPC desktop flow, CLI command, local URL, staging URL, or documented user journey. It tests behavior, captures evidence, and does not modify code.

Gstack Cso guides security reviews of OPC code using OWASP and STRIDE lenses. Use it to inspect Electron IPC, local bridges, secrets, plugins, command execution, and file access.

Gstack Investigate guides systematic debugging for bugs, crashes, regressions, flaky behavior, and provider issues. It emphasizes reproduction, evidence, and the smallest fix before verification.

Gstack Review performs a pre-merge review of the current diff to catch regressions, security issues, missing tests, and cleanup problems. Use it before shipping or asking whether a change is safe.

Hook Development explains how to create Claude Code plugin hooks for events like PreToolUse, PostToolUse, Stop, and SessionStart. Use it to validate tool use, block unsafe commands, and run event-driven automation with prompt-based or command hooks.

MCP Integration helps configure Model Context Protocol servers in Claude Code plugins so they can use external tools and services. It covers .mcp.json, inline mcpServers, transport types, auth, and tool naming.

Plugin Settings documents the `.claude/plugin-name.local.md` pattern for per-project plugin configuration and state. Use it when reading YAML frontmatter and markdown content to control plugin behavior from hooks, commands, or agents.

Plugin Structure explains the Claude Code plugin directory layout, manifest setup, and auto-discovered components. Use it when creating or organizing plugins, adding commands, agents, skills, hooks, or configuring paths with ${CLAUDE_PLUGIN_ROOT}.

Skill Development helps you create and improve Claude Code skills, including their structure, metadata, and progressive disclosure. Use it when writing a new skill, refining skill descriptions, or organizing skill content.

Writing Hookify Rules explains how to create Hookify markdown rules with YAML frontmatter, patterns, conditions, and messages. Use it when setting up or editing rules that watch Claude Code events like bash, file, stop, or prompt.