Skill 1
Judge rubric writer
Use when an AI judge needs a written task definition, pass criteria, forced-reject criteria, review scale, and minimum evidence requirement before scoring workflow output.
Back to Skill Library
Agent governance workflow library
AI judge review gates
Review rubrics, evidence boundaries, bias probes, disagreements, and authority before an AI judge scores workflow output.
This is a complete workflow library with 5 individual skills. Download the full library or pick the specific skill folder your team needs first.
Individual skills in this library
Use one skill at a time, or keep the full workflow together.
Some AI tools expect one skill folder per upload. Download the full library when you want the whole workflow, or download an individual skill when you only need one job done.
Skill 2
Judge evidence boundary mapper
Use when an AI judge needs allowed evidence, blocked evidence, sensitive data classes, source trust, and redaction rules before it reads workflow output or source material.
Skill 3
Bias probe set builder
Use when an AI judge needs known-good, known-bad, order-swap, longer-worse, style-only, missing-evidence, wrong-tool-argument, or prompt-injection probes before being trusted.
Skill 4
Judge disagreement router
Use when an AI judge conflicts with a human reviewer, another judge, a known label, a source trace, an eval run, or a policy rule.
Skill 5
Judge authority gatekeeper
Use when a team needs to decide whether an AI judge may assist, triage, recommend, approve with human review, or remain blocked.
Security fit check
Is the public AI judge review gates library enough, or does this need deeper review?
Use the public library when the workflow is low-risk, the inputs are already sanitized, and a team member can review the output before it reaches a buyer or customer.
Do deeper review when this workflow touches real tools, data sources, role ownership, approval paths, or customer-facing output.
AI evaluationAI OperationsSecurityWorkflow Owner
Good deeper-review trigger signals
- The workflow touches customer, prospect, CRM, proposal, security, pricing, or campaign data.
- Different teams disagree on the approved source of truth.
- The AI output could become customer-facing, revenue-impacting, or compliance-sensitive.
- You need reusable eval checks before asking more people to use the workflow.