{"slug": "anthropics-claude-certified-architect-exam-cca-f-the-agent-that-did-exactly-what", "title": "Anthropic’s Claude Certified Architect Exam (CCA-F): The Agent That Did Exactly What the Injected…", "summary": "Anthropic's Claude Certified Architect Exam (CCA-F) Part 7 argues that AI agent safety is an architecture problem, not a prompting problem, and that real safety lives at the tool boundary with a deny, ask, or allow gate. The article covers five ideas for hardening agents: grounding against hallucination, prompt-injection defense with a deterministic gate, a hardcoded-over-softcoded trust hierarchy, human-in-the-loop approval gates, and red-team-evals-fallback discipline.", "body_md": "Member-only story\n\n# Anthropic’s Claude Certified Architect Exam (CCA-F): The Agent That Did Exactly What the Injected Text Told It To\n\n## CCA-F Part 7: Why AI agent safety is an architecture problem, not a prompting problem, and how that single shift in thinking is the judgment a Claude certification scenario actually tests\n\n*You can spend a week word-smithing the perfect “please refuse malicious instructions” system prompt and still watch one line of untrusted text walk your agent into a destructive action. Real safety lives at the tool boundary, in a deny, ask, or allow gate that ranks the operator over the user over anything the model merely reads.*\n\nIn this article:You will learn why the safest answer to an AI agent safety question is almost never “improve the prompt.” We cover the five ideas that turn a fragile agent into a hardened one: grounding against confident hallucination, prompt-injection defense with a deterministic gate, the hardcoded-over-softcoded trust hierarchy, human-in-the-loop approval gates, and the red-team-evals-fallback discipline that measures quality…", "url": "https://wpnews.pro/news/anthropics-claude-certified-architect-exam-cca-f-the-agent-that-did-exactly-what", "canonical_source": "https://pub.towardsai.net/anthropics-claude-certified-architect-exam-cca-f-the-agent-that-did-exactly-what-the-injected-8285d3135afd?source=rss----98111c9905da---4", "published_at": "2026-07-28 06:36:51+00:00", "updated_at": "2026-07-28 07:12:09.085956+00:00", "lang": "en", "topics": ["ai-safety", "ai-agents", "ai-research"], "entities": ["Anthropic", "Claude"], "alternates": {"html": "https://wpnews.pro/news/anthropics-claude-certified-architect-exam-cca-f-the-agent-that-did-exactly-what", "markdown": "https://wpnews.pro/news/anthropics-claude-certified-architect-exam-cca-f-the-agent-that-did-exactly-what.md", "text": "https://wpnews.pro/news/anthropics-claude-certified-architect-exam-cca-f-the-agent-that-did-exactly-what.txt", "jsonld": "https://wpnews.pro/news/anthropics-claude-certified-architect-exam-cca-f-the-agent-that-did-exactly-what.jsonld"}}