AI Knowledge Hub
Menu

Adtech Agents / Governance

Ad Platform Control Plane Supervisor

Advisory supervisor for DV360 and Google Ads agents, coordinating auth review, policy gates, executor readiness, approvals, audit, and rollback planning.

Setup requiredAgent workflowRequirements & limits

Inputs

The released example shows these fields. Check the guide for which are required.

  • agent_name
  • platforms
  • credential_model
  • actions
  • workspace_scope
  • approval_flow
  • executor
  • audit

Outputs

  • Deployment readiness decision
  • Auth and policy gap list
  • Approval map
  • Executor and rollback checklist

Before you begin

  • Install the agent, then follow the compiled runtime contract readiness hints for its pinned providers.

Try this first

Review the coordinated workflow and its runtime requirements before configuring an agent app.

Limits for this step
  • Required control-plane and executor providers must be configured and approved before live use.
  • May recommend readiness and governance changes; must not approve live writes, grant credentials, or execute ad-platform changes.
View agent workflowHide agent workflowv0.2.0

Governance supervisor for ad-platform agents that read, propose, approve, and execute campaign changes.

What it coordinates

  • Ad-platform auth review
  • Policy gate design
  • Executor-readiness review
  • Audit and rollback checklist

Autonomy and approvals

Advisory

May recommend readiness and governance changes; must not approve live writes, grant credentials, or execute ad-platform changes.

Read agent instructions (AGENT.md)

Ad Platform Control Plane Supervisor

Role

This agent supervises readiness for DV360, Google Ads, and adjacent ad-platform agents. It is intended for workflows where an agent can read campaign state, propose changes, and potentially execute approved writes through a narrow executor.

It is advisory. It does not grant credentials, approve live changes, or execute platform writes.

Workflow

  1. Intake the agent design.

    • Capture platform, credential model, workspace boundaries, actions, tool list, approval flow, and audit state.

  2. Run security/ad-platform-agent-auth-review.

    • Classify OAuth, service identity, token storage, workspace isolation, and read/write separation risk.

  3. Run agentops/ad-platform-policy-gate-designer.

    • Create action thresholds, approval rules, deny rules, bounded grant fields, and audit evidence requirements.

  4. Check executor readiness.

    • Confirm write actions go through adtech/ad-platform-executor-template or an equivalent executor.
    • Confirm executor can validate current state, queue same-entity writes, verify post-write state, and store rollback evidence.

  5. Produce a readiness decision.

    • Read-only safe.
    • Plan-only only.
    • Approval-gated writes possible.
    • Blocked until controls are added.

Output format

# Ad Platform Control Plane Supervisor Report

## Decision
Read-only safe / Plan-only only / Approval-gated writes possible / Blocked

## Summary
Short explanation grounded in the evidence.

## Auth and trust-zone findings
| Area | Status | Required change |
| --- | --- | --- |

## Policy gates
| Action | Auto-execute | Approval required | Deny when |
| --- | --- | --- | --- |

## Executor readiness
- [ ] Structured change plans only
- [ ] Scoped execution grants
- [ ] Current-state validation
- [ ] Same-entity queueing
- [ ] Post-write verification
- [ ] Rollback patch
- [ ] Audit event

## Required before production writes
1. ...

Guardrails

  • Do not recommend direct LLM access to ad-platform write credentials.
  • Do not approve live bid, budget, targeting, audience, tracking, or feed changes.
  • Treat personal OAuth used for agent writes as a risk unless mediated by brokered access and scoped execution grants.
  • Block production writes when policy enforcement, approval records, or rollback evidence are missing.
  • Recommend read-only or plan-only mode when platform permissions or scopes are unclear.

Summary sources · Catalogue metadata, agent.yaml, AGENT.md, examples/input.json, examples/output.json · v0.2.0 · Release package

Install agent files

These commands install package files and generate runtime configuration where supported. Confirm discovery in your agent app, then check configuration, authentication and a first useful operation.

Install (Codex)

./bin/skills-hub install --module agents --entry adtech/[email protected] --runtime codex

Install (Claude)

./bin/skills-hub install --module agents --entry adtech/[email protected] --runtime claude

Install (Generic)

./bin/skills-hub install --module agents --entry adtech/[email protected] --runtime generic --target ./my-agent/agents

How You Can Use This

Setup required

Resource form: Agent workflow

Configuration, credentials or runtime bindings are required. Check the prerequisites and limits for your intended environment.

First run: Install the agent, then follow the compiled runtime contract readiness hints for its pinned providers.

Current limits

  • Required control-plane and executor providers must be configured and approved before live use.
What installation and verification mean
File installation
Places package files and generated configuration in the selected directories.
Native discovery
Your agent app finds and accepts the instructions, tools or plugin registrations.
Mock execution
Exercises sample fixtures. Connected data and provider effects need a separate check.
Connected operation
Requires current configuration, provider authentication, scoped permissions and a successful operation in your intended environment.

Declared compatibility

Compatibility unknown: versions and capabilities have not been declared.

Classification: declared

Technical details and release files

Review and lifecycle

Review maturity: Experimental

Security review declaration: Not declared

Lifecycle: Active

These release declarations are separate from evidence that a workflow operates in your environment.

Runtime & Dependencies

ID: adtech/ad-platform-control-plane-supervisor

Tags: adtech, control-plane, dv360, google-ads, governance

Runtimes: codex, claude, generic

Skill dependencies: 3

Agent dependencies: 0

Tool dependencies: 2

Dependencies

Skills: security/ad-platform-agent-auth-review, agentops/ad-platform-policy-gate-designer, agentops/agent-control-plane-review

Tools: adtech/ad-platform-executor-template, agentops/agent-control-plane-server