Fix Related Concepts link labels and clean up Typst remnants (#4)

Siddhant-K-code · ona-agent · web-flow · commit 33f02e2b42ae · 2026-03-04T16:20:47.000Z
- Fix 27 Related links where labels didn't match target chapters
  (subsection refs like 'Agent Threat Model' pointing to wrong chapters,
  self-references, workflow labels pointing to Ch 23/24 generically)
- Fix Typst syntax leak in 17-agent-loop.mdx image reference
- Fix garbled description in appendix-b-glossary.mdx

Co-authored-by: Ona &lt;no-reply@ona.com&gt;
diff --git a/src/content/chapters/04-context-windows.mdx b/src/content/chapters/04-context-windows.mdx
@@ -164,4 +164,4 @@ as a first-class discipline - with dedicated ownership, monitoring, and
 continuous improvement - consistently get better results from the same
 models than teams that treat it as an afterthought.
 
-**Related Workflows:** [Setting Up Distill for Monorepo Context](/23-first-agent-production/) **Related Practices:** [Context Budget Policy](/24-security-checklist/) 
+**Related Workflows:** [Your First Agent in Production](/23-first-agent-production/) **Related Practices:** [Security Checklist & Review Policies](/24-security-checklist/) 
diff --git a/src/content/chapters/05-context-engineering-stack.mdx b/src/content/chapters/05-context-engineering-stack.mdx
@@ -237,4 +237,4 @@ history, preferences, and current intent. The key challenge is privacy -
 customer data must be handled according to your data protection
 policies, and PII must be redacted before it enters the model’s context.
 
-**Related Concepts:** [Context Window](/04-context-windows/), [Token Budget](/04-context-windows/) **Related Workflows:** [Implementing Meta-MCP for Token Reduction](/23-first-agent-production/) 
+**Related Concepts:** [Context Window](/04-context-windows/), [Context Windows](/04-context-windows/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/) 
diff --git a/src/content/chapters/06-rag-vs-agentic-search.mdx b/src/content/chapters/06-rag-vs-agentic-search.mdx
@@ -157,7 +157,7 @@ search, git history search), returns structured results with metadata
 can ask follow-up questions based on initial results). This investment
 pays off as your agents become more sophisticated.
 
-**Related Concepts:** [Context Window](/04-context-windows/), [Context Engineering Stack](/05-context-engineering-stack/) **Related Practices:** [Building a Codebase Knowledge Base](/22-agent-maturity-model/)
+**Related Concepts:** [Context Window](/04-context-windows/), [Context Engineering Stack](/05-context-engineering-stack/) **Related Practices:** [AGENTS.md & Codebase Onboarding](/13-agents-md-onboarding/)
 
 
 > *"Your AI agent has more permissions than your senior developers."*
diff --git a/src/content/chapters/08-zanzibar-for-agents.mdx b/src/content/chapters/08-zanzibar-for-agents.mdx
@@ -170,7 +170,7 @@ database schema to generate meaningful test data. These permissions
 should be added incrementally as needs are identified, not granted
 preemptively "just in case."
 
-**Related Concepts:** [Agent Identity](/07-agent-security-crisis/), [Delegation Chains](/08-zanzibar-for-agents/) **Related Workflows:** [Implementing Agent Authorization with OpenFGA](/23-first-agent-production/)
+**Related Concepts:** [The Agent Security Crisis](/07-agent-security-crisis/), [Zanzibar for AI Agents](/08-zanzibar-for-agents/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/)
 
 ## OPA vs. OpenFGA: Choosing your policy engine
 Two authorization systems dominate enterprise discussions for agent
diff --git a/src/content/chapters/09-prompt-injection.mdx b/src/content/chapters/09-prompt-injection.mdx
@@ -175,4 +175,4 @@ on them to catch sophisticated ones.
 - [ ] Security team reviews agent traces monthly for novel attack patterns
 
 
-**Related Concepts:** [Agent Security Crisis](/07-agent-security-crisis/), [Sandboxing](/10-sandboxing/) **Related Practices:** [Agent Threat Model](/24-security-checklist/) 
+**Related Concepts:** [Agent Security Crisis](/07-agent-security-crisis/), [Sandboxing](/10-sandboxing/) **Related Practices:** [Security Checklist & Review Policies](/24-security-checklist/) 
diff --git a/src/content/chapters/10-sandboxing.mdx b/src/content/chapters/10-sandboxing.mdx
@@ -131,7 +131,7 @@ on having the right policies. Each layer catches what the previous
 layers miss. The goal isn’t perfect security - it’s making the cost of a
 successful attack high enough that the risk is manageable.
 
-**Related Concepts:** [Zanzibar](/08-zanzibar-for-agents/), [Prompt Injection](/09-prompt-injection/) **Related Workflows:** [Security Checklist for Agent Deployment](/24-security-checklist/) **Related Tools:** Anthropic sandbox-runtime, OpenFGA, AgentGuard
+**Related Concepts:** [Zanzibar](/08-zanzibar-for-agents/), [Prompt Injection](/09-prompt-injection/) **Related Workflows:** [Security Checklist & Review Policies](/24-security-checklist/) **Related Tools:** Anthropic sandbox-runtime, OpenFGA, AgentGuard
 
 
 > *"Standards enable ecosystems. The wild west phase is ending."*
diff --git a/src/content/chapters/11-mcp.mdx b/src/content/chapters/11-mcp.mdx
@@ -179,4 +179,4 @@ waits, and eventually gives up or works around the missing tool. Monitor
 your MCP servers the same way you monitor your production APIs: uptime,
 latency, error rate, and throughput.
 
-**Related Concepts:** [Context Window](/04-context-windows/), [Meta-MCP](/05-context-engineering-stack/) **Related Workflows:** [Setting Up MCP Servers for Your Team](/23-first-agent-production/) 
+**Related Concepts:** [Context Window](/04-context-windows/), [The Context Engineering Stack](/05-context-engineering-stack/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/) 
diff --git a/src/content/chapters/12-a2a-communication.mdx b/src/content/chapters/12-a2a-communication.mdx
@@ -128,4 +128,4 @@ specific requirements they address.
 - [ ] Health checks verify downstream agent availability
 
 
-**Related Concepts:** [MCP](/11-mcp/), [Multi-Agent Systems](/18-multi-agent-systems/) **Related Workflows:** [Agent-to-Agent Communication Setup](/23-first-agent-production/) 
+**Related Concepts:** [MCP](/11-mcp/), [Multi-Agent Systems](/18-multi-agent-systems/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/) 
diff --git a/src/content/chapters/13-agents-md-onboarding.mdx b/src/content/chapters/13-agents-md-onboarding.mdx
@@ -185,7 +185,7 @@ How do you know if your AGENTS.md is working?
 If your agent consistently violates conventions or fails tests, your
 AGENTS.md needs improvement.
 
-**Related Concepts:** [Context Engineering](/05-context-engineering-stack/), [Codebase Knowledge Graph](/06-rag-vs-agentic-search/) **Related Workflows:** [Onboarding Agents to Your Codebase](/23-first-agent-production/)
+**Related Concepts:** [Context Engineering](/05-context-engineering-stack/), [RAG vs. Agentic Search](/06-rag-vs-agentic-search/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/)
 
 ## GitHub agentic workflows
 In February 2026, GitHub announced Agentic Workflows - automated,
@@ -443,7 +443,7 @@ first-attempt success rates and lower review burden on human engineers -
 each layer addresses a different failure mode, and gaps in any layer
 force the agent to guess.
 
-**Related Concepts:** [AGENTS.md](/13-agents-md-onboarding/), [Context Engineering](/05-context-engineering-stack/), [MCP](/11-mcp/) **Related Workflows:** [Onboarding Agents to Your Codebase](/23-first-agent-production/), [Security Checklist](/24-security-checklist/)
+**Related Concepts:** [AGENTS.md](/13-agents-md-onboarding/), [Context Engineering](/05-context-engineering-stack/), [MCP](/11-mcp/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/), [Security Checklist](/24-security-checklist/)
 
 
 > *"When an agent fails, teams ask 'What happened?' and have no good
diff --git a/src/content/chapters/15-cost-tracking.mdx b/src/content/chapters/15-cost-tracking.mdx
@@ -153,4 +153,4 @@ change and comprehensive evals on a schedule.
 - [ ] Context accumulation is managed (summarization, pruning)
 
 
-**Related Concepts:** [Token Budget](/04-context-windows/), [Model Routing](/15-cost-tracking/) **Related Practices:** [Agent Cost Management](/24-security-checklist/) 
+**Related Concepts:** [Context Windows](/04-context-windows/), [Model Selection & Routing](/31-model-selection/) **Related Practices:** [Agent Cost Control & FinOps](/28-agent-cost-control/) 
diff --git a/src/content/chapters/16-incident-response.mdx b/src/content/chapters/16-incident-response.mdx
@@ -141,7 +141,7 @@ problem and keeps trying increasingly expensive approaches. Each retry
 adds to the context window, which increases the cost of subsequent
 retries. The fix is per-session cost limits with automatic termination.
 
-**Related Concepts:** [Agent Traces](/14-agent-traces/), [Security](/07-agent-security-crisis/) **Related Practices:** [Incident Response Plan](/24-security-checklist/), [Backpressure](/32-backpressure/)
+**Related Concepts:** [Agent Traces](/14-agent-traces/), [Security](/07-agent-security-crisis/) **Related Practices:** [Incident Response for Agent Failures](/16-incident-response/), [Backpressure](/32-backpressure/)
 
 
 > *"Agent orchestration is distributed systems engineering applied to
diff --git a/src/content/chapters/17-agent-loop.mdx b/src/content/chapters/17-agent-loop.mdx
@@ -18,8 +18,7 @@ executes that tool, appends the result to the conversation, and sends
 everything back to the model. This continues until the model produces a
 final answer or hits a limit.
 
-![](/images/fig-agent-loop.png).],
-) 
+![The agent loop: think, act, observe, repeat.](/images/fig-agent-loop.png)
 
 The simplicity of this loop is deceptive. The engineering challenge
 isn’t the loop itself - it’s everything around it: how you manage the
@@ -190,4 +189,4 @@ approval, runs it in staging, waits for approval, and only then proposes
 the production migration. The gate configuration should be part of the
 task specification, not hardcoded in the agent framework.
 
-**Related Concepts:** [Agent Security](/07-agent-security-crisis/), [Observability](/14-agent-traces/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/) 
+**Related Concepts:** [Agent Security](/07-agent-security-crisis/), [Agent Traces & OpenTelemetry](/14-agent-traces/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/) 
diff --git a/src/content/chapters/18-multi-agent-systems.mdx b/src/content/chapters/18-multi-agent-systems.mdx
@@ -183,4 +183,4 @@ see individual service logs, but you can’t reconstruct the end-to-end
 flow. Invest in trace propagation before deploying multi-agent systems
 to production.
 
-**Related Concepts:** [A2A Protocol](/12-a2a-communication/), [Agent Loop](/17-agent-loop/) **Related Workflows:** [Setting Up Multi-Agent Workflows](/23-first-agent-production/) 
+**Related Concepts:** [Agent-to-Agent Communication (A2A)](/12-a2a-communication/), [Agent Loop](/17-agent-loop/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/) 
diff --git a/src/content/chapters/19-agent-memory.mdx b/src/content/chapters/19-agent-memory.mdx
@@ -152,7 +152,7 @@ performance.
 - [ ] Agent can answer "why was this changed?" for recent changes
 
 
-**Related Concepts:** [Context Engineering](/05-context-engineering-stack/), [Knowledge Graph](/06-rag-vs-agentic-search/) **Related Workflows:** [Creating a Codebase Knowledge Graph](/23-first-agent-production/)
+**Related Concepts:** [Context Engineering](/05-context-engineering-stack/), [RAG vs. Agentic Search](/06-rag-vs-agentic-search/) **Related Workflows:** [Your First Agent in Production](/23-first-agent-production/)
 
 
 > *"The hardest problems in agentic engineering aren’t technical.
diff --git a/src/content/chapters/20-ai-fatigue.mdx b/src/content/chapters/20-ai-fatigue.mdx
@@ -227,4 +227,4 @@ with deployment scripts and tuning guidance.
 - ☐ Team has a shared channel for agent wins and failures
 - ☐ Agent scope is limited to tasks with clear verification criteria
 
-**Related Concepts:** [Review Burden](/20-ai-fatigue/), [The Conductor Model](/21-conductor-model/), [Backpressure](/32-backpressure/) **Related Practices:** [Introducing Agents to a Skeptical Team](/24-security-checklist/) 
+**Related Concepts:** [AI Fatigue & Sustainable Adoption](/20-ai-fatigue/), [The Conductor Model](/21-conductor-model/), [Backpressure](/32-backpressure/) **Related Practices:** [The Agent Adoption Playbook](/33-adoption-playbook/) 
diff --git a/src/content/chapters/appendix-b-glossary.mdx b/src/content/chapters/appendix-b-glossary.mdx
@@ -1,6 +1,6 @@
 ---
 title: "Appendix B: Glossary"
-description: "[Google’s global authorization system. Basis for OpenFGA and similar systems.], ) ]"
+description: "Key terms and definitions used throughout the book."
 order: 201
 ---
 

Original file line number	Diff line number	Diff line change
`@@ -175,4 +175,4 @@ on them to catch sophisticated ones.`
`175`	`175`	`- [ ] Security team reviews agent traces monthly for novel attack patterns`
`176`	`176`
`177`	`177`
`178`		`-Related Concepts: [Agent Security Crisis](/07-agent-security-crisis/), [Sandboxing](/10-sandboxing/) Related Practices: [Agent Threat Model](/24-security-checklist/)`
	`178`	`+Related Concepts: [Agent Security Crisis](/07-agent-security-crisis/), [Sandboxing](/10-sandboxing/) Related Practices: [Security Checklist & Review Policies](/24-security-checklist/)`
Original file line number	Diff line number	Diff line change
`@@ -128,4 +128,4 @@ specific requirements they address.`
`128`	`128`	`- [ ] Health checks verify downstream agent availability`
`129`	`129`
`130`	`130`
`131`		`-Related Concepts: [MCP](/11-mcp/), [Multi-Agent Systems](/18-multi-agent-systems/) Related Workflows: [Agent-to-Agent Communication Setup](/23-first-agent-production/)`
	`131`	`+Related Concepts: [MCP](/11-mcp/), [Multi-Agent Systems](/18-multi-agent-systems/) Related Workflows: [Your First Agent in Production](/23-first-agent-production/)`
Original file line number	Diff line number	Diff line change
`@@ -153,4 +153,4 @@ change and comprehensive evals on a schedule.`
`153`	`153`	`- [ ] Context accumulation is managed (summarization, pruning)`
`154`	`154`
`155`	`155`
`156`		`-Related Concepts: [Token Budget](/04-context-windows/), [Model Routing](/15-cost-tracking/) Related Practices: [Agent Cost Management](/24-security-checklist/)`
	`156`	`+Related Concepts: [Context Windows](/04-context-windows/), [Model Selection & Routing](/31-model-selection/) Related Practices: [Agent Cost Control & FinOps](/28-agent-cost-control/)`