Backlinks!

Signed-off-by: teocns <teocns@gmail.com>

Author: teocns

docs/agents/creating-agents.md

@@ -4,6 +4,10 @@ description: "How to create agents: frontmatter fields, system prompt design, mo
 sidebar_position: 2
 ---
 
+:::tip[Official reference]
+For how sub-agents work in Claude Code, see the [official sub-agents docs](https://code.claude.com/docs/en/sub-agents).
+:::
+
 ## File Location
 
 Place agent files in the `agents/` directory of your plugin:

docs/index.mdx

@@ -15,7 +15,11 @@ hero:
       variant: minimal
 ---
 
-import { LinkCard, CardGrid } from '@astrojs/starlight/components';
+import { LinkCard, CardGrid, Aside } from '@astrojs/starlight/components';
+
+<Aside type="tip">
+New to Claude Code? Start with the [official quickstart](https://code.claude.com/docs/en/quickstart) to get installed and running. This guide picks up where that leaves off.
+</Aside>
 
 <CardGrid>
   <LinkCard

docs/patterns/agentic-engineering.md

@@ -83,3 +83,4 @@ The agent doesn't know what it doesn't know. Your job is to fill the gaps with c
 
 - [Workflow patterns](workflow-patterns.md) — concrete patterns for common tasks
 - [Common pitfalls](../reference/common-pitfalls.md) — mistakes to avoid
+- [Official best practices](https://code.claude.com/docs/en/best-practices) — Anthropic's own recommendations

docs/patterns/context-management.md

@@ -4,6 +4,10 @@ description: "How to keep Claude Code effective across long sessions and large c
 sidebar_position: 3
 ---
 
+:::tip[Official reference]
+For CLAUDE.md mechanics and memory configuration, see the [official memory docs](https://code.claude.com/docs/en/memory). For all configuration options, see [settings](https://code.claude.com/docs/en/settings).
+:::
+
 How to keep Claude Code effective across long sessions and large codebases.
 
 ## The Context Window

docs/patterns/workflow-patterns.md

@@ -78,3 +78,7 @@ When you're new to a codebase or unfamiliar module.
 - **Review diffs, not just the outcome.** Claude may change more than necessary — check what actually changed.
 - **Use `/compact` between tasks.** Frees context for the next piece of work.
 - **Don't fight the agent.** If Claude is going in circles, step back and reframe the task rather than repeating the same instruction.
+
+:::tip[Want more?]
+The official docs have a broader set of [common workflows](https://code.claude.com/docs/en/common-workflows) covering PR creation, CI integration, and more.
+:::

docs/reference/common-pitfalls.md

@@ -62,3 +62,6 @@ Mistakes engineers make when starting with Claude Code — and how to avoid them
 
 **The fix:** One task at a time. Commit between tasks. Each change should be reviewable on its own.
 
+## Keep Learning
+
+For more on working effectively with Claude Code, see the [official best practices](https://code.claude.com/docs/en/best-practices).

docs/reference/links.md

@@ -8,10 +8,19 @@ Curated links for engineers using Claude Code. We cover the advanced setup and p
 
 ## Official Documentation
 
-- [Claude Code Documentation](https://code.claude.com/docs) — Start here for installation, settings reference, and getting started
-- [Claude Code Docs (Anthropic)](https://docs.anthropic.com/en/docs/claude-code/overview) — Anthropic's reference docs
+This guide covers the opinionated setup. The official docs cover everything else.
+
+- [Claude Code Documentation](https://code.claude.com/docs) — The starting point for everything
+- [Quickstart](https://code.claude.com/docs/en/quickstart) — Installation and first session
+- [Settings Reference](https://code.claude.com/docs/en/settings) — All configuration options
+- [Hooks](https://code.claude.com/docs/en/hooks) — Full hooks API and event types
+- [MCP](https://code.claude.com/docs/en/mcp) — Model Context Protocol reference
+- [Skills](https://code.claude.com/docs/en/skills) — How skills work under the hood
+- [Sub-agents](https://code.claude.com/docs/en/sub-agents) — Agent spawning and management
+- [Best Practices](https://code.claude.com/docs/en/best-practices) — Anthropic's recommendations
+- [CLI Reference](https://code.claude.com/docs/en/cli-reference) — All CLI commands and flags
 - [Claude Code GitHub](https://github.com/anthropics/claude-code) — Source, issues, and discussions
-- [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) — The protocol for connecting Claude to external tools
+- [Model Context Protocol](https://modelcontextprotocol.io/) — The MCP specification
 - [Anthropic Cookbook](https://github.com/anthropics/anthropic-cookbook) — Practical examples and patterns
 
 ## Skills Ecosystem

docs/setup/claude-md.md

@@ -4,6 +4,10 @@ description: "How to write effective CLAUDE.md files — the highest-leverage co
 sidebar_position: 1
 ---
 
+:::tip[New to CLAUDE.md?]
+Start with the [official CLAUDE.md guide](https://code.claude.com/docs/en/memory) for the basics. This page is the opinionated blueprint for making yours great.
+:::
+
 CLAUDE.md is the highest-leverage configuration surface for Claude Code. It's loaded into every conversation, so every line must earn its place.
 
 ## What CLAUDE.md Does

docs/setup/hooks-playbook.md

@@ -4,9 +4,11 @@ description: "Essential hooks every project should have — enforcing what CLAUD
 sidebar_position: 3
 ---
 
-CLAUDE.md is advisory — Claude follows it because it's in context, but nothing enforces compliance. Hooks are code. They execute automatically on specific events and can block, modify, or extend Claude's behavior.
+Here's the thing about CLAUDE.md: it's a suggestion. Claude *usually* follows it, but nothing enforces compliance. Hooks are different. Hooks are code — they execute automatically on specific events and can block, modify, or extend Claude's behavior. If CLAUDE.md is a polite request, hooks are a locked door.
 
-For the full hooks reference, see the [official docs](https://code.claude.com/docs/hooks).
+:::tip[Official reference]
+For the full hooks API, event types, and configuration options, see the [official hooks docs](https://code.claude.com/docs/en/hooks). This page gives you the five hooks worth setting up first.
+:::
 
 ## The Essential Five
 

docs/setup/mcp-wiring.md

@@ -4,9 +4,11 @@ description: "Connect Claude Code to your external tools — GitHub, databases,
 sidebar_position: 4
 ---
 
-MCP (Model Context Protocol) connects Claude Code to external tools and data sources. Instead of copy-pasting context from GitHub issues, database queries, or Jira tickets, Claude can access them directly.
+This is where Claude Code goes from "smart editor" to "team member with access." MCP (Model Context Protocol) connects Claude to your external tools and data sources — GitHub, databases, project management, Slack, and more. No more copy-pasting context. Claude reads the issue, queries the schema, writes the code, and opens the PR.
 
-For the full MCP reference, see the [official docs](https://code.claude.com/docs/mcp).
+:::tip[Official reference]
+For the full MCP API, authentication flows, and server registry, see the [official MCP docs](https://code.claude.com/docs/en/mcp). This page covers the three connections worth setting up first.
+:::
 
 ## The Three Most Impactful Connections