Skip to content
ReviewRamble

The weekly brief

AI, reviewed, in your inbox.

  • New reviews of tools, models, MCP servers and repos, with the verdict up front
  • Re-checked prices and ratings when something changes
  • New analysis pieces as they are published

Models, agents, MCP servers and the tools actually worth your time.

One email a week. Unsubscribe with the link in any brief. How we use your email

Context engineering

Context engineering is the practice of choosing and maintaining everything that goes into a language model's context window during a task: instructions, retrieved documents, tool definitions, tool results and conversation history.

By Divya Sharma · Checked Nov 15, 2026 · How we review

A model only knows what is in its training data and what is in front of it on each call. Context engineering is about the second part. It asks, at every step, which information the model needs now and what can be left out.

Why it matters

Agents run in loops. Each tool call, file read or search result adds tokens, and the input grows with every turn. Anthropic's engineering team argues that context should be treated as a finite resource with diminishing returns: as the input grows, models get worse at recalling what is in it. That decline is often called context rot. More context is not automatically better context.

Common techniques

- Keeping the system prompt short and specific. - Retrieving only relevant passages, as in retrieval-augmented generation, instead of pasting whole documents. - Summarising or trimming old turns, known as context compaction. - Storing notes outside the window and reloading them when needed, a form of agent memory. - Handing narrow jobs to subagents so that their working detail stays out of the main thread.

How it differs from prompt engineering

Prompt engineering is mostly a one-time task: write a good instruction. Context engineering is repeated on every turn, because the set of possible information keeps changing as the agent works. Anthropic describes it as the natural progression of prompt engineering.

Example

A coding agent asked to fix a failing test does better if it is given the test, the error and the function under test than if it is given the whole repository. The skill is in picking that small, relevant set and refreshing it as the work moves on.

Related terms

Prompt engineering
Prompt engineering is the practice of writing, structuring and testing the instructions and examples given to a language model so that it produces the output an application needs, without changing the model's weights.
Agent memory
Agent memory is the set of techniques that let an AI agent keep information beyond a single context window, such as facts about a user or past work, by storing it outside the model and loading it back when needed.
Retrieval-augmented generation
Retrieval-augmented generation (RAG) is a pattern where a system first searches a collection of documents for passages relevant to a question, then gives those passages to a language model so its answer draws on them.
Context window
A context window is the maximum amount of text, counted in tokens, that a language model can consider at once, covering the system prompt, conversation, documents, tool results and the response it writes.
Context rot
Context rot is the drop in a language model's accuracy and reliability as its input grows longer, even when the input still fits inside the context window and the needed information is present.
Context compaction
Context compaction is the step where an agent summarises a conversation that is nearing its context window limit and continues in a fresh window that starts from that summary, so a long task can keep going.

Sources