← Agent catalogue
Learning governance (global promotion)

Preview portrait

Vision / platform

Learning governance (global promotion)

Observing governance lane for tenant-local learning candidates and promotion evidence packets.

In buildCatalogue status
ObservingRuntime posture

Tenant learning governance posture plus explicit promotion packets; approve/revoke never auto-promotes cross-tenant.

  • Learning governance
  • Observing
  • Promotion packets
  • No auto-promote
Open consoleBook a Demo

Expanded capabilities

  • Expose tenant learning-governance posture and emitter coverage
  • Host review/apply for learning candidates with human gates
  • Issue promotion packets with evidence requirements
  • Revoke candidates without cross-tenant side effects
  • Block Global ATLAS auto-push on apply paths

How they learn & improve

This lane *is* the self-improve control plane: every agent’s RECORD_ONLY candidates converge here for human judgment.

  1. Collect candidates from siloed emitters and posture pins.
  2. Require approver roles to review/apply with optional retention targets.
  3. Record revoke/apply audit so improvement stays attributable.

How they work together

Learning governance stitches FEEDBACK, domain agents, MNEMOSYNE/ATLAS knowledge, and Global ATLAS park/review.

Practical benefits

  • One inbox for agent lessonsOperators stop hunting per-module AI settings — improvements are reviewed in one governed queue.
  • Apply with eyes openApply confirms proposal kind/target and optional retentionUntil before anything changes.
  • Kill a bad lessonRevoke removes a candidate without silent re-application.

Learning and improvement stay human-visible: candidates are RECORD_ONLY until an operator reviews and applies them. Sensitive sends, publishes, and approvals never run silently.

Related agents

Learning governance (global promotion) — AgentIQ Agent Catalogue · AgentIQ Labs