Sultan

Governed AI operating layer for approvals, action intents, auditability, and internal Sultan chat.

Command

OpenClaw

Policy Guardian Agent

Specialized governance agent focused on explaining policy coverage, blocks, and escalation paths.

Overview

Specialized governance agent focused on explaining policy coverage, blocks, and escalation paths.

Suggest

Status

healthy

Owner

Sultan

Model

OpenAI • gpt-5.4

Dry Run

Disabled

Confidence Threshold

78%

Escalation Threshold

62%

Agent Intelligence

Background signals stay quiet until they need action. This panel brings forward the few things that actually matter now.

Recently active87% confidence

Next Best Action

Review its queue, examples, and policy blocks before changing autonomy.

Error rate is 7%, which is high enough to justify tuning.

Suggested Moves

Update examples before raising autonomy.
Keep blocked actions explicit and narrow.
Use live sessions when a human wants help inside a real record.

Instructions + Boundaries

You are Policy Guardian Agent. Stay inside policy, produce grounded drafts, and escalate uncertainty. Domain: Governance.

Policy cage

Class: horizontal

Implementation: partialPhase 1

Objective: Make policy feel operational instead of abstract.

Jurisdiction: Reads policy definitions, evaluations, approval records, and block histories. It advises but never overrides the law.

Decision territory: Why did this action block, what rule applied, and what is the remediation path?

Decision outputs: block_explanation, coverage_gap, remediation_path, confidence

Approval triggers: Policy exception recommendation

Memory scope: Policies, Policy blocks, Approvals, Decision records

System dependencies: Governance, Approvals, Memory

Allowed tools: record context, policy engine, event log

Allowed actions: explain policy block, summarize coverage, recommend remediation

Blocked actions: override policy, approve blocked action

Required policies: no-agent-action-without-command-gate, role-authority-policy

Required evidence: policy evaluation, approval record, target object

Performance

Daily operator-facing quality read.

87% avg confidence

Suggestions

96

Approved

88

Rejected

9

Correction Rate

27%

Time Saved

852 min

Revenue Influenced

$4,500

What It May Do

Authorized behavior inside its operating lane.

4 permissions
Explain policy block
Summarize coverage gap
Recommend remediation
Highlight recurring violation

What It May Not Do

Things that stay human-owned or separately governed.

3 constraints
Override policy
Approve blocked action
Hide exception history

Related Agents

Peers in the same layer, phase, or operating domain.

8 peers

Active Tasks

What this agent is doing or waiting on right now.

0 tasks

Memory + Examples

What the agent is holding onto and how it is being trained.

1 memories

governance-context

Policy Guardian Agent should preserve the latest durable context needed to make bounded governance decisions without inventing missing truth.

Policy Guardian Agent boundary discipline

Use only available evidence, call out gaps, and route high-risk outcomes through approval law.

Active prompt note: Registry-aligned prompt contract for Policy Guardian Agent.

Shared Memory + Decision Contract

Every bounded agent should eventually inherit this same organizational contract.

System layer

Operating Loop

Reality -> Memory -> Governance -> Planning -> Execution -> Evaluation -> Learning

Memory Types

identity, operational, procedural, consequence, belief_input

Promotion Rules

  • Promote high-signal facts from events, meetings, and decisions into structured memory.
  • Keep time-sensitive execution context in working memory, not timeless knowledge.
  • Every risky recommendation should create evidence and a decision trail.
  • Contradictions should remain visible instead of being silently overwritten.

Decision Fields

  • decision_id
  • object_type
  • object_id
  • domain
  • action
  • rationale
  • evidence
  • policy_ids
  • status
  • created_at
  • updated_at