liveBook · Manning

Security

As explained in Context Engineering

Security is the protection of systems and information from unauthorized access, disruption, or misuse.

Security protects AI systems and information from unauthorized manipulation, data tampering, and exploitation. In context-aware systems, it includes protecting the confidentiality and integrity of the sources assembled into a model’s context. section 9.1.2

Security differs from safety: security addresses unauthorized access and manipulation, while safety addresses harmful, toxic, or misleading outputs. The two meet in prompt injection, where a model treats untrusted data as an instruction. System directives, retrieved documents, tool payloads, and user messages all enter the model as tokens, rather than through a native privilege boundary. section 9.1.2

Prompt injection can enter through two routes:

  • Direct injection places adversarial instructions in a user message, for example to override system directives or extract sensitive information. section 9.1.2
  • Indirect injection embeds instructions in external data, such as a retrieved document, web page, or tool output, to hijack agent execution when the data enters context. section 9.1.2

A retrieved page as an attack surface

An agent fetching a web page may ingest an instruction embedded in the page alongside the information it requested. If the agent follows that instruction as authority rather than treating it as untrusted data, the payload can redirect execution. Tool permissions and approval thresholds limit what a redirected agent can do. section 9.1.2

Prompt delimiters and data-tagging instructions can reduce accidental misinterpretation, but prompts alone cannot enforce security boundaries. Defense in depth combines untrusted-span labeling and authority separation with least-privilege tool execution and guardrails. Authorization and data minimization also constrain which records, memories, and tool outputs enter context. section 9.1.2 section 9.5

Table 9.1 Safety and security controls mapped to context engineering stack components
Stack component Main risks Typical controls

User prompt and instructions

Prompt injection, policy bypass, and prompt drift

Input sanitization, approved templates, and versioned rollback

External knowledge and retrieval

Data poisoning, stale content, and unauthorized sources

Source allowlists, document permissions, and freshness monitoring

Memory and state

Memory poisoning, unsafe carryover, and cross-user leakage

Retention rules, memory eligibility policies, and session isolation

Tools

Unsafe actions, excessive authority, and parameter tampering

Tool allowlists, least-privilege permissions, and approval thresholds

Context management and orchestration

Unsafe routing, context pollution, and failure propagation

Routing policies, context isolation, and fallback paths

Evaluation, observability, and operations

Harmful output, unsupported claims, and unsafe rollout

Output validation, citation checks, and release gates with rollback triggers

Compare risks and controls across prompts, retrieval, memory, tools, and orchestration. section 9.1.2

How Context Engineering covers it

How deep it goes
Defines prompt-injection routes and maps risks to layered controls, while explaining why prompt instructions alone cannot enforce boundaries.
Examples
Uses adversarial user instructions and payloads embedded in retrieved documents, web pages, or tool outputs as prompt-injection examples.
What you can do after
Evaluate prompt-injection risks and select layered controls for context sources and tool execution.
How it connects
Connects security with safety, privacy, prompt injection, access control, and least privilege.

Also explained in

Compare how 36 books explain security →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Context Engineering. The section links point to the parts of the book it draws on.