Documentation Index

Fetch the complete documentation index at: https://docs.aifabrix.ai/llms.txt

Use this file to discover all available pages before exploring further.

Skill levels and promotion

Prev Next

Role assistants become more reliable through evidence from completed work. Skill levels make that maturity visible — without granting new authority.

Why it matters

Enterprises need to see which workers are proven useful for a role and task type. Skill levels communicate reliability and adoption readiness for executives and operators. They must not be confused with permission escalation.

Promotion answers: “Is this worker consistently helpful for this role?” — not “Should this worker have more system access?”

How it works

Trainee → Capable → Trusted → Expert

Promotion considers signals such as:

  • completed tasks and successful outcomes
  • confirmed human impact (accepted vs corrected results)
  • repeated reliability on recurring task types
  • data quality improvements surfaced through evidence

Signals come from Evidence Fabrix — aggregated from completed work, not from chat volume or token usage.

Promotion does not:

  • increase permissions or dimensions
  • bypass approvals or policy
  • override governance or certification
  • grant direct API access

It improves transparency and trust in the worker's assistance quality.

Limits

Skill snapshots, promotion thresholds, and Role Assistant dashboards depend on Evidence Fabrix features enabled in your environment. Treat promotion as a visibility signal until your deployment certifies Role Assistant evidence as live.

Example

Sales Assistant promoted from Capable to Trusted.

This means repeated useful pipeline reviews and measurable impact — not permission to update CRM without approval or to see customers outside regional dimensions.

Business value

Visible Role Assistant improvement, safer adoption conversations, and alignment between learning and accountability.