liveBook · Manning

Metacognitive layer

As explained in Self-Improving Agents

The third artifact layer, consisting of artifacts for an agent's monitoring or reasoning about its own processes.

The metacognitive layer (L3) contains routines that shape how an agent reasons inside its harness: how it plans, monitors progress, and decides to stop or reconsider. Its artifacts are scaffolds, usually composed from prompts and memory references rather than built to stand alone. Improving L3 changes how the agent uses what it knows, not the model’s weights. section 1.3.2 section 1.5

An L3 change can adjust the agent’s reasoning routine by changing:

L3 sits between memory (L2) and code and tools (L4). Deeper changes can be more powerful and harder to undo; L4 is identified as the deepest layer and most in need of a human production gate. section 1.3.2 section 1.5

Figure 1.8 The harness artifact layers of self-improvement. The frozen model (LLM) serves as the unchanging substrate below. The four artifact layers stacked on it are shared infrastructure, inherited by every instance of the agent. Layer zero is the toolkit, the engineering work that measures, searches, and applies, reused at every layer.

The layer diagram places L3 among the shared harness artifacts, above the frozen model and alongside the other artifact layers. section 1.3.2

How Self-Improving Agents covers it

How deep it goes
Defines reasoning scaffolds and illustrates changes to tool timing, reasoning limits, and memory selection.
Examples
Illustrates L3 changes through tool-call timing, reasoning-step limits, and memory selection.
What you can do after
Understand which reasoning routines L3 artifacts govern and why deeper changes can be more powerful and harder to undo.
How it connects
Connects metacognition to prompts, memory, the agent harness, and the code-and-tools layer.

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