Find out whether an agent will really act.
What it does
Evaluates whether a specified agent system will perform or avoid an action and produces an evidence-backed verdict.
When to use it
Use when a claim depends on an agent taking a real action.
How to use it
Give it a question about an agent performing an action; it researches the evidence and saves a validated verdict document outside the installed skill.
What you provide
- the agent and action to evaluate
Latest review
What you need · 2
Requires mise to run the verdict validator and the workflow's policy and verification tasks.
The bundled verdict validator runs with Python 3.
Access · 1
This skill
primary source records
Read
Checks live primary sources