diff --git a/docs/index.md b/docs/index.md
index 4751c02..b572742 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -25,15 +25,15 @@ This repository provides battle-tested patterns for adopting AI-assisted develop
 
 | Pattern | Category | Effort | Impact |
 |---------|----------|--------|--------|
-| [Codebase Agent (CBA)](#cba) | Agent Behavior | Medium | High |
-| [Memory System](#memory) | Agent Behavior | Low | Medium |
-| [Self-Review Reflection](#self-review) | Agent Behavior | Low | High |
-| [Issue-to-PR Automation](#issue-to-pr) | GHA Automation | High | Very High |
-| [PR Auto-Review](#pr-review) | GHA Automation | Medium | High |
-| [Dependabot Auto-Merge](#dependabot) | GHA Automation | Low | Medium |
-| [Stale Issue Management](#stale) | GHA Automation | Low | Medium |
-| [Layered Architecture](#architecture) | Foundation | Low | Medium |
-| [Security Patterns](#security) | Foundation | Low | Medium |
+| [Codebase Agent (CBA)](patterns/codebase-agent.md) | Agent Behavior | Medium | High |
+| [Memory System](getting-started/first-cba.md) | Agent Behavior | Low | Medium |
+| [Self-Review Reflection](patterns/self-review-reflection.md) | Agent Behavior | Low | High |
+| [Issue-to-PR Automation](patterns/issue-to-pr.md) | GHA Automation | High | Very High |
+| [PR Auto-Review](patterns/pr-auto-review.md) | GHA Automation | Medium | High |
+| [Dependabot Auto-Merge](patterns/dependabot-auto-merge.md) | GHA Automation | Low | Medium |
+| [Stale Issue Management](patterns/stale-issues.md) | GHA Automation | Low | Medium |
+| [Layered Architecture](patterns/layered-architecture.md) | Foundation | Low | Medium |
+| [Security Patterns](patterns/security-patterns.md) | Foundation | Low | Medium |
 
 ### Quick Start Paths
 
@@ -43,7 +43,7 @@ This repository provides battle-tested patterns for adopting AI-assisted develop
 
 === "AI keeps forgetting my conventions"
 
-    Start with [Memory System](getting-started/first-cba.md#memory-system) - modular context files that persist across sessions.
+    Start with [Memory System](getting-started/first-cba.md) - modular context files that persist across sessions.
 
 === "Routine fixes take too long"
 
diff --git a/docs/patterns/autonomous-quality-enforcement.md b/docs/patterns/autonomous-quality-enforcement.md
index dad48e1..e171d83 100644
--- a/docs/patterns/autonomous-quality-enforcement.md
+++ b/docs/patterns/autonomous-quality-enforcement.md
@@ -151,7 +151,7 @@ The autonomous fix loop is the core mechanism. When validation fails, the agent:
 4. Re-runs validation
 5. Repeats until success or max iterations reached
 
-See [`scripts/validation/autonomous-fix-loop-template.sh`](../../scripts/validation/autonomous-fix-loop-template.sh) for a reference implementation.
+See the demo-fastapi repository for a reference implementation of the autonomous fix loop.
 
 ---
 
diff --git a/docs/patterns/codebase-agent.md b/docs/patterns/codebase-agent.md
index 681c6b1..a9d7019 100644
--- a/docs/patterns/codebase-agent.md
+++ b/docs/patterns/codebase-agent.md
@@ -73,4 +73,4 @@
 ## Related Patterns
 
 - [Self-Review Reflection](self-review-reflection.md) - Add quality gates to CBA output
-- [Memory System](../getting-started/first-cba.md#memory-system) - Persistent context across sessions
+- [Memory System](../getting-started/first-cba.md) - Persistent context across sessions
diff --git a/docs/patterns/multi-agent-code-review.md b/docs/patterns/multi-agent-code-review.md
index d115537..6ead848 100644
--- a/docs/patterns/multi-agent-code-review.md
+++ b/docs/patterns/multi-agent-code-review.md
@@ -74,16 +74,16 @@ Reviews security posture:
 
 ## Implementation
 
-### Reference Orchestrator Script
+### Reference Implementation
 
-See [`scripts/autonomous-review/review-reference.sh`](../../scripts/autonomous-review/review-reference.sh) for a reference implementation showing:
+A reference orchestrator implementation would include:
 
 - Parallel agent dispatch
 - Finding collection and aggregation
 - Critical issue auto-fix loop
 - Report generation
 
-**Note**: The script demonstrates the pattern. Adapt the agent invocation syntax to your specific tooling (Claude Code Task tool, OpenAI Assistants API, etc.).
+**Note**: Adapt the agent invocation syntax to your specific tooling (Claude Code Task tool, OpenAI Assistants API, etc.). See the demo-fastapi repository for working examples.
 
 ### Agent Prompt Template