Patterns (#19559)

* Patterns section + links

* One page and use <GitHubCode/> component

* Add remaining images

* fix typo

* fix typo

* Images

* fixes

* Remove unneeded images

* links

* PCX Review

---------

Co-authored-by: Jun Lee <[email protected]>

Author: 2 people

src/assets/images/agents/01-prompt-chaining.jpg

@@ -0,0 +1,88 @@
+---
+pcx_content_type: concept
+title: Patterns
+sidebar:
+  order: 3
+head: []
+---
+
+import { GitHubCode } from "~/components";
+
+This page lists and defines common patterns for implementing AI agents, based on [Anthropic's patterns for building effective agents](https://www.anthropic.com/research/building-effective-agents).
+
+Code samples use the [AI SDK](https://sdk.vercel.ai/docs/foundations/agents), running in [Durable Objects](/durable-objects).
+
+## Prompt Chaining
+
+Decomposes tasks into a sequence of steps, where each LLM call processes the output of the previous one.
+
+![Figure 1: Prompt Chaining](~/assets/images/agents/01-prompt-chaining.jpg)
+
+[**Interactive Demo**](https://ai.cloudflare.com/anthropic-agent-patterns)
+
+<GitHubCode
+	repo="cloudflare/agents"
+	file="guides/anthropic-patterns/src/flows/01 sequential.txt"
+	commit="56fb801dc8f980f080b36fcdbd94005099829495"
+	lang="ts"
+/>
+
+## Routing
+
+Classifies input and directs it to specialized followup tasks, allowing for separation of concerns.
+
+![Figure 2: Routing](~/assets/images/agents/2_Routing.jpg)
+
+[**Interactive Demo**](https://ai.cloudflare.com/anthropic-agent-patterns)
+
+<GitHubCode
+	repo="cloudflare/agents"
+	file="guides/anthropic-patterns/src/flows/02 routing.txt"
+	commit="56fb801dc8f980f080b36fcdbd94005099829495"
+	lang="ts"
+/>
+
+## Parallelization
+
+Enables simultaneous task processing through sectioning or voting mechanisms.
+
+![Figure 3: Parallelization](~/assets/images/agents/3_Parallelization.jpg)
+
+[**Interactive Demo**](https://ai.cloudflare.com/anthropic-agent-patterns)
+
+<GitHubCode
+	repo="cloudflare/agents"
+	file="guides/anthropic-patterns/src/flows/03 parallel.txt"
+	commit="56fb801dc8f980f080b36fcdbd94005099829495"
+	lang="ts"
+/>
+
+## Orchestrator-Workers
+
+A central LLM dynamically breaks down tasks, delegates to Worker LLMs, and synthesizes results.
+
+![Figure 4: Orchestrator Workers](~/assets/images/agents/4_Orchestrator-Workers.jpg)
+
+[**Interactive Demo**](https://ai.cloudflare.com/anthropic-agent-patterns)
+
+<GitHubCode
+	repo="cloudflare/agents"
+	file="guides/anthropic-patterns/src/flows/04 orchestrator.txt"
+	commit="56fb801dc8f980f080b36fcdbd94005099829495"
+	lang="ts"
+/>
+
+## Evaluator-Optimizer
+
+One LLM generates responses while another provides evaluation and feedback in a loop.
+
+![Figure 5: Evaluator-Optimizer](~/assets/images/agents/5_Evaluator-Optimizer.jpg)
+
+[**Interactive Demo**](https://ai.cloudflare.com/anthropic-agent-patterns)
+
+<GitHubCode
+	repo="cloudflare/agents"
+	file="guides/anthropic-patterns/src/flows/05 evaluator.txt"
+	commit="56fb801dc8f980f080b36fcdbd94005099829495"
+	lang="ts"
+/>