Knowledge Flywheel Monitor FAQs

Question 1

What are the four 'Golden Signals' tracked by the skill?

Accepted Answer

The signals are Velocity Trend (is learning increasing?), Citation Pipeline (are citations useful?), Research Closure (is research being utilized?), and Reuse Concentration (is the whole pool active?).

Question 2

Do I need to install external tools to use this?

Accepted Answer

The skill works best with the 'ao' CLI installed, but it includes a robust fallback mode that can compute health metrics directly from your .agents/ directory using standard shell commands.

Question 3

Where can I see the results of the health check?

Accepted Answer

The skill generates a detailed Markdown report at .agents/flywheel-status.md, which includes pool depths, velocity metrics, and specific recommendations for improving agent performance.

Question 4

What is the 'Flywheel' model in this context?

Accepted Answer

The Flywheel is a cycle where agent sessions generate transcripts that are processed into learnings and patterns. These artifacts are then promoted to a knowledge pool where future sessions can find and use them, creating a compounding feedback loop.

Question 5

How does this skill identify 'friction' in development?

Accepted Answer

Friction is identified by monitoring bottlenecks like high rates of uncited learnings, stagnant velocity trends, or a high ratio of orphaned research that never results in actionable patterns.

Knowledge Flywheel Monitor

Knowledge Flywheel Monitor

Key Features

Use Cases

Key Features

Use Cases