Enable javascript in your browser for better experience. Need to know to enable it? Go here.
Close
Ask Tai
Ask Tai
Download

Harness quality evaluation

Techniques Back
Published : Nov 04, 2026
Nov 2026
Assess ?

Nowadays, organizations and teams come up with custom harnesses tailored to their projects, including agent skills, plugins, AGENTS.md and deterministic guardrails. Harness quality evaluation has become an important technique. Guardrails and harnesses help a coding agent course-correct and improve overall quality, but only if they actually work.

A harness usually includes both deterministic and non-deterministic components. For non-deterministic harnesses, our teams use PluginEval to evaluate coding agent plugins or skills. For deterministic harnesses, we introduce a test suite to enable TDD-style development of any guardrails, ensuring each one actually works and providing evidence of its effectiveness. We recommend running evaluations on a scheduled basis (such as nightly or alongside CI).

Related blips

Download the PDF

 

 

 

English | Português

Sign up for the Technology Radar newsletter

 

 

Subscribe now

Visit our archive to read previous volumes

Go to archive

Thoughtworks acknowledges the Traditional Owners of the land where we work and live, and their continued connection to Country. We pay our respects to Elders past and present. Aboriginal and Torres Strait Islander peoples were the world's first scientists, technologists, engineers and mathematicians. We celebrate the stories, culture and traditions of Aboriginal and Torres Strait Islander Elders of all communities who also work and live on this land. 

 

As a company, we invite Thoughtworkers to be actively engaged in advancing reconciliation and strengthen their solidarity with the First Peoples of Australia. Since 2019, we have been working with Reconciliation Australia to formalize our commitment and take meaningful action to advance reconciliation. We invite you to review our Reconciliation Action Plan.