← Back to Patterns

Capability Confidence

Availablecapability graph•advanced

Overview

Surface confidence intervals and evidence weight — Fourth Law applies to capability truth as well as model output.

Why this matters

  • Explicit uncertainty beats fake precision
  • Confidence informs routing, not human ranking
  • Low confidence → escalate or require stronger evidence
  • Try it

    >
    SDK:

    Use cases

  • High-risk task routing
  • Evidence tier upgrades
  • Audit explanations for match decisions
  • See Also

  • Verify capability
  • Capability routing
  • Same route in the other SDKs

    The TypeScript tab above is the contract. These calls hit that same route.

    >
    SDK: