Add Property Governance Protocol for AdCP 3.0

.changeset/property-governance-protocol.md

@@ -0,0 +1,9 @@
+---
+"adcontextprotocol": minor
+---
+
+Add Property Governance Protocol support to get_products
+
+- Add optional `property_list` parameter to get_products request for filtering products by property list
+- Add `property_list_applied` response field to indicate whether filtering was applied
+- Enables buyers to pass property lists from governance agents to sales agents for compliant inventory discovery

docs.json

@@ -80,13 +80,6 @@
               "docs/protocols/envelope-examples"
             ]
           },
-          {
-            "group": "adagents.json",
-            "pages": [
-              "docs/media-buy/capability-discovery/adagents",
-              "docs/media-buy/capability-discovery/authorized-properties"
-            ]
-          },
           {
             "group": "Signals Protocol",
             "pages": [
@@ -101,6 +94,36 @@
               }
             ]
           },
+          {
+            "group": "Governance Protocol [3.0]",
+            "pages": [
+              "docs/governance/index",
+              {
+                "group": "Property Governance",
+                "pages": [
+                  "docs/governance/property/index",
+                  "docs/governance/property/specification",
+                  "docs/governance/property/authorized-properties",
+                  "docs/governance/property/adagents",
+                  {
+                    "group": "Tasks",
+                    "pages": [
+                      "docs/governance/property/tasks/index",
+                      "docs/governance/property/tasks/list_property_features",
+                      "docs/governance/property/tasks/property_lists",
+                      "docs/governance/property/tasks/validate_property_delivery"
+                    ]
+                  }
+                ]
+              },
+              {
+                "group": "Brand Standards",
+                "pages": [
+                  "docs/governance/brand-standards/index"
+                ]
+              }
+            ]
+          },
           {
             "group": "Curation Protocol",
             "pages": [

docs/governance/brand-standards/index.mdx

@@ -0,0 +1,90 @@
+---
+sidebar_position: 1
+title: Brand Standards
+---
+
+<Info>
+**AdCP 3.0 Proposal** - This protocol is under development for AdCP 3.0. Feedback welcome via [GitHub Discussions](https://github.com/adcontextprotocol/adcp/discussions).
+</Info>
+
+Brand Standards governance enables buyers to declare brand requirements that governance agents use to automatically apply appropriate rules across advertising workflows.
+
+## Overview
+
+Brand Standards addresses three concerns:
+
+| Concern | Description | Mechanism |
+|---------|-------------|-----------|
+| **Brand Identity** | Who is the advertiser? | Brand manifest declarations |
+| **Creative Guidelines** | What can the brand show? | Creative constraints and policies |
+| **Audience Constraints** | Who should see the ads? | Industry-specific targeting rules |
+
+## Brand Manifest
+
+The brand manifest is the core declaration that buyers provide to governance agents:
+
+```json
+{
+  "brand_manifest": {
+    "brand_name": "ToyBrand",
+    "industry": "toys",
+    "target_audience": "children_under_13",
+    "prohibited_categories": ["gambling", "alcohol", "tobacco"],
+    "creative_guidelines": {
+      "max_animation_duration": 15,
+      "require_accessibility": true
+    }
+  }
+}
+```
+
+When governance agents receive a brand manifest, they automatically apply:
+
+- **Regulatory compliance**: COPPA for children's brands, GDPR consent requirements
+- **Content filtering**: Block categories inappropriate for the brand's audience
+- **Creative validation**: Enforce brand-specific creative guidelines
+
+## Integration with Property Governance
+
+Brand manifests work alongside property lists. When creating a property list, include the brand manifest to get automatic rule application:
+
+```json
+{
+  "tool": "create_property_list",
+  "arguments": {
+    "name": "ToyBrand Q1 Properties",
+    "base_properties": [
+      {
+        "selection_type": "publisher_tags",
+        "publisher_domain": "kidsmedia.com",
+        "tags": ["kids_family", "educational"]
+      }
+    ],
+    "filters": {
+      "countries_all": ["US"],
+      "channels_any": ["display", "video"],
+      "feature_requirements": [
+        { "feature_id": "brand_safety", "min_value": 90, "max_value": 100 }
+      ]
+    },
+    "brand_manifest": {
+      "brand_name": "ToyBrand",
+      "industry": "toys",
+      "target_audience": "children_under_13"
+    }
+  }
+}
+```
+
+The governance agent uses the brand manifest to apply COPPA requirements, filter out non-child-safe properties, and enforce relevant industry rules.
+
+## Coming Soon
+
+Detailed specification and task definitions for Brand Standards governance are under development. Key planned features:
+
+- **Brand manifest validation**: Verify manifest completeness and consistency
+- **Industry rule sets**: Pre-defined rules for common industries (toys, alcohol, pharmaceuticals, finance)
+- **Creative compliance**: Validate creatives against brand guidelines
+- **Audience verification**: Ensure targeting complies with brand audience constraints
+
+See the [Governance Protocol overview](/docs/governance/index) for the broader context of how Brand Standards fits into AdCP governance.

docs/governance/index.mdx

@@ -0,0 +1,187 @@
+---
+sidebar_position: 1
+title: Governance Protocol
+---
+
+<Warning>
+**Draft for AdCP 3.0** - These governance protocols are under active development by the AAO Governance Working Group. Feedback welcome via [GitHub Discussions](https://github.com/adcontextprotocol/adcp/discussions).
+</Warning>
+
+The Governance Protocol provides standardized mechanisms for compliance, brand safety, and quality control across advertising workflows. It enables specialized agents to evaluate, filter, and score advertising entities against configurable criteria.
+
+## Mission
+
+> Define and operate transparent, privacy-safe, brand-suitable campaign governance that encodes human heuristics in open protocol objects, making campaigns easy to launch, safe to scale, and measurable against verifiable outcomes for buyers, sellers, and auditors.
+
+## Protocol Domains
+
+The Governance Protocol covers four distinct governance domains:
+
+| Domain | What It Governs | Key Mechanisms |
+|--------|-----------------|----------------|
+| **[Property Governance](/docs/governance/property/index)** | Where ads can run | Property lists, compliance filtering, adagents.json authorization |
+| **[Brand Standards](/docs/governance/brand-standards/index)** | Brand requirements | Brand manifests, creative guidelines, audience constraints |
+| **Creative Governance** | What creatives are compliant | Format validation, content moderation, accessibility |
+| **Campaign Governance** | What can be bought | Budget controls, approval workflows, policy compliance |
+
+Each domain has specialized governance agents that provide data, filtering, and scoring capabilities.
+
+## Working Group Areas
+
+The AAO Governance Working Group is developing standards across six areas that inform these protocols:
+
+| Area | Focus | Protocol Impact |
+|------|-------|-----------------|
+| Brand Safety & Suitability | Safe, appropriate ad contexts | Property Governance, Content filtering |
+| Brand Manifesto & Standards Library | Machine-readable brand policies | Brand Standards |
+| Creative Standards & Validation | Technical and ethical creative criteria | Creative Governance |
+| Process & Human Oversight | Human-in-the-loop checkpoints | All protocols |
+| Regulatory & Compliance | Legal and regional frameworks | All protocols |
+| Measurement & Verification | Proving compliance | All protocols |
+
+## Architecture
+
+Governance agents operate across the **full campaign lifecycle**:
+
+- **Setup time**: Configure property lists, brand requirements, and compliance rules
+- **Real-time**: Influence bid-time decisions through cached rules and scoring
+- **Post-bid**: Measurement, verification, and reporting on compliance
+
+```
+┌─────────────────────────────────────────────────────────────────────────┐
+│                           BUYER AGENT                                    │
+│  - Aggregates governance from specialized agents                         │
+│  - Issues auth_tokens for sellers to access governance data              │
+│  - Source of truth for compliant lists and brand requirements            │
+└───────────────────────────┬─────────────────────────────────────────────┘
+                            │ create_property_list, webhooks
+        ┌───────────────────┼───────────────────┐
+        ▼                   ▼                   ▼
+┌───────────────┐   ┌───────────────┐   ┌───────────────┐
+│ Consent Agent │   │ Scope3 Agent  │   │ Brand Safety  │
+│ (Compliance)  │   │               │   │    Agent      │
+├───────────────┤   ├───────────────┤   ├───────────────┤
+│ consent_qual  │   │ carbon_score  │   │ content_cat   │
+│ tcf_version   │   │ climate_risk  │   │ brand_risk    │
+│ coppa_cert    │   │ green_host    │   │ sentiment     │
+└───────────────┘   └───────────────┘   └───────────────┘
+                            │
+                            │ get_property_list (with auth_token)
+                            ▼
+┌─────────────────────────────────────────────────────────────────────────┐
+│                         SELLER AGENT (DSP/SSP)                          │
+│  - Caches resolved property lists for bid-time decisions                 │
+│  - Receives updates via webhooks when lists change                       │
+└─────────────────────────────────────────────────────────────────────────┘
+```
+
+## Common Patterns
+
+### Feature Discovery
+
+All governance agents advertise their capabilities via discovery tasks:
+
+```json
+{
+  "tool": "list_property_features",
+  "arguments": {}
+}
+```
+
+Response:
+```json
+{
+  "features": [
+    { "feature_id": "consent_quality", "type": "quantitative", "range": { "min": 0, "max": 100 } },
+    { "feature_id": "coppa_certified", "type": "binary" },
+    { "feature_id": "carbon_score", "type": "quantitative", "range": { "min": 0, "max": 100 } }
+  ]
+}
+```
+
+### Brand Manifest
+
+Buyers declare their brand identity, and governance agents automatically apply appropriate rules:
+
+```json
+{
+  "brand_manifest": {
+    "brand_name": "ToyBrand",
+    "industry": "toys",
+    "target_audience": "children_under_13"
+  }
+}
+```
+
+When a governance agent sees this manifest, it automatically applies COPPA requirements, content filtering for children's brands, and other industry-specific rules.
+
+### Prompt-Based Policies
+
+Governance protocols support **natural language prompts** for policy definitions rather than rigid keyword lists:
+
+```json
+{
+  "policy": "Avoid content about violence, controversial politics, or adult themes. Sports news is excellent. Entertainment is generally acceptable.",
+  "exclusions_prompt": "Block content containing hate speech, illegal activities, or ongoing litigation against our company."
+}
+```
+
+This enables AI-powered verification agents to understand context and nuance.
+
+### Webhook Notifications
+
+Governance agents notify subscribers when evaluations change:
+
+```json
+{
+  "webhook_url": "https://buyer.example.com/webhooks/list-changed",
+  "events": ["list_updated", "property_removed", "score_changed"]
+}
+```
+
+## Integration with Media Buy
+
+The Media Buy Protocol consumes governance data at multiple stages:
+
+- **Product discovery**: Pass `property_list_ref` to `get_products` to filter inventory
+- **Media buy creation**: Reference property lists to constrain where ads can run
+- **Authorization**: adagents.json validates agent authority to sell properties
+
+```json
+{
+  "tool": "get_products",
+  "arguments": {
+    "brief": "UK video inventory for Q1",
+    "property_list_ref": {
+      "agent_url": "https://buyer-agent.example.com",
+      "list_id": "pl_q1_uk_premium",
+      "auth_token": "..."
+    }
+  }
+}
+```
+
+## Getting Started
+
+**Buyers:**
+1. Subscribe to governance agents for property data and brand safety
+2. Create property lists with filters and brand manifests
+3. Aggregate results into final compliant lists
+4. Share references with sellers (with auth tokens)
+
+**Governance Agent Implementers:**
+1. Implement feature discovery tasks (`list_property_features`, etc.)
+2. Implement stateful list management (CRUD operations)
+3. Support webhooks to notify when evaluations change
+4. See the [Property Protocol Specification](/docs/governance/property/specification) for detailed implementation guidance
+
+## Protocol Sections
+
+<CardGroup cols={2}>
+  <Card title="Property Governance" icon="building" href="/docs/governance/property/index">
+    Control where ads can run with property lists, compliance filtering, and publisher authorization via adagents.json.
+  </Card>
+  <Card title="Brand Standards" icon="shield-check" href="/docs/governance/brand-standards/index">
+    Define brand requirements including creative guidelines, audience constraints, and industry-specific rules.
+  </Card>
+</CardGroup>

docs/media-buy/capability-discovery/adagents.mdx → docs/governance/property/adagents.mdx

@@ -1,9 +1,9 @@
 ---
-sidebar_position: 8
+sidebar_position: 3
 title: adagents.json Tech Spec
 ---
 
-The `adagents.json` file provides a standardized way for publishers to declare which sales agents are authorized to sell their advertising inventory. This specification addresses authorization transparency and helps prevent unauthorized reselling of publisher inventory.
+The `adagents.json` file provides a standardized way for publishers to declare their properties and authorize sales agents. This is the foundation of Property Governance - it defines what properties exist and who can sell them.
 
 <Tip>
 **[AdAgents.json Builder](https://adcontextprotocol.org/adagents)** - Validate existing files or create new ones with guided validation

docs/media-buy/capability-discovery/authorized-properties.mdx → docs/governance/property/authorized-properties.mdx

@@ -1,5 +1,6 @@
 ---
-title: Authorized Properties
+sidebar_position: 2.5
+title: Understanding Authorization
 description: Understanding property authorization in AdCP - preventing unauthorized resale and validating sales agent relationships
 keywords: [authorized properties, adagents.json, unauthorized resale, ads.txt, sales agent authorization, property validation]
 ---
@@ -268,11 +269,11 @@ For complete technical details on implementing the `adagents.json` file format,
 - Validation code examples and error handling
 - Security considerations and best practices
 
-See the **[adagents.json Tech Spec](/docs/media-buy/capability-discovery/adagents)** for complete implementation guidance.
+See the **[adagents.json Tech Spec](/docs/governance/property/adagents)** for complete implementation guidance.
 
 ## Related Documentation
 
 - **[`list_authorized_properties`](/docs/media-buy/task-reference/list_authorized_properties)** - API reference for property discovery
 - **[Product Discovery](/docs/media-buy/product-discovery/)** - How authorization integrates with product discovery
 - **[Properties Schema](https://adcontextprotocol.org/schemas/v2/core/property.json)** - Technical property data model
-- **[adagents.json Tech Spec](/docs/media-buy/capability-discovery/adagents)** - Complete `adagents.json` implementation guide
+- **[adagents.json Tech Spec](/docs/governance/property/adagents)** - Complete `adagents.json` implementation guide