Enterprise AI Measurement Guide
Agent Effectiveness
AI Fluency Score Summary
What specific engineering behaviors and infrastructure elements make up our AI Fluency Score, and which ones should we invest in developing to improve agent effectiveness?
What it shows
The AI Fluency summary block shows the organization's aggregate AI Fluency Score for the engineering team alongside the input counts that build it: the number of engineers measured, total app sessions, Practices adopted (structured workflow behaviors Larridin tracks), Skills demonstrated, MCP Servers connected, and Plugins in use. This gives the Fluency Score context. A score of 72 built from 3 practices and 1 MCP server is very different from 72 built from 8 practices and 6 MCP servers.
Why it matters
The AI Fluency Score without its components is an output number that tells leaders how the org rates but not why, or what to change. The component breakdown is the diagnostic layer: a low Practices count means engineers have not developed structured workflows for working with agents; low MCP Servers means the agent stack is not connected to the tools where work actually happens; a low Skills count means engineers are not demonstrating the range of agent interaction behaviors associated with high Outcome Success. Each component points to a specific intervention.
The Larridin angle
Larridin's Fluency measurement is built from observable behaviors rather than self-reported confidence, which means it reflects what engineers actually do with agents, not what they think they are doing. The component breakdown makes that behavioral evidence actionable for CHROs building training programs and CTOs evaluating team capability.