Emergent Capability: Agents Do What You Didn’t Design For

The Idea

When tools are atomic and parity is complete, the agent can accomplish things you never explicitly designed. A user asks: “Cross-reference my meeting notes with my task list and tell me what I’ve committed to but haven’t scheduled.” You didn’t build a commitment tracker. But if the agent can read notes and tasks, it can compose them into the outcome.

The flywheel that makes this self-reinforcing:

  1. Build with atomic tools and parity.
  2. Users ask for things you didn’t anticipate.
  3. Agent composes tools to accomplish them — or fails, revealing a gap.
  4. You observe patterns in what’s being requested.
  5. Add domain tools or prompts to make common patterns efficient.
  6. Repeat.

Why It Matters

Emergent capability changes how you build products. You’re not trying to imagine every feature upfront. You’re creating a capable foundation and learning from what emerges. This reveals latent demand without surveys, interviews, or guesswork — the requests themselves are the signal.

The Test

Can the agent handle open-ended requests in your domain that you didn’t explicitly anticipate? If yes, you’ve earned emergence. If no, your tools are too coarse, your parity is incomplete, or both.