Framework

Three waves: prompt, context, harness

The question it answers

When a firm puts AI models to work, where should its leaders invest effort: writing prompts, engineering context, or building the harness around the model?

Origin
Boonstra, n.d.; Anthropic, 2025; Lopopolo, 2026
Diagram type
Process by phase
Level of analysis
Firm
Topics
AI for leadersStrategy & leadership
Glass sphere with a glowing neural-network core, icon for AI for leaders

Diagram

Three waves: prompt, context, harness

  1. Three waves
  2. Prompt engineeringHow to write the request.
  3. Context engineeringChoosing what information the model sees.
  4. Harness engineeringBuilding the environment, tools, constraints and feedback loops so the agent works reliably.

Pitfalls

Pitfall

The formula agent = model + harness is a shorthand and should not be quoted as if it were a direct quotation.

Sources

References

Listed A–Z by author surname, Harvard style. A source with a DOI links to https://doi.org/…; one without links to its page or says plainly that it has no DOI.

  1. Anthropic (2025) Effective context engineering for AI agents. Anthropic Research.No DOIanthropic.comSource page
  2. Boonstra, L. (n.d.) Prompt engineering [whitepaper].No DOIkaggle.comSource page
  3. Lopopolo, R. (2026) Harness engineering: leveraging Codex in an agent-first world. OpenAI.No DOIopenai.comSource page