feat(governance): Add get_creative_features task for creative governance

.changeset/creative-governance.md

@@ -0,0 +1,14 @@
+---
+"adcontextprotocol": minor
+---
+
+Add `get_creative_features` task for creative governance
+
+Introduces the creative analog of `get_property_features` — a general-purpose task for evaluating creatives and returning feature values. Supports security scanning, creative quality assessment, content categorization, and any other creative evaluation through the same feature-based pattern used by property governance.
+
+New schemas:
+- `get-creative-features-request.json` — accepts a creative manifest and optional feature_ids filter
+- `get-creative-features-response.json` — returns feature results with discriminated union (success/error)
+- `creative-feature-result.json` — individual feature evaluation (value, confidence, expires_at, etc.)
+
+Also adds `creative_features` to the governance section of `get_adcp_capabilities` response, allowing agents to advertise which creative features they can evaluate.

docs.json

@@ -143,6 +143,13 @@
                     ]
                   }
                 ]
+              },
+              {
+                "group": "Creative Governance",
+                "pages": [
+                  "docs/governance/creative/index",
+                  "docs/governance/creative/get_creative_features"
+                ]
               }
             ]
           },
@@ -593,6 +600,13 @@
                     ]
                   }
                 ]
+              },
+              {
+                "group": "Creative Governance",
+                "pages": [
+                  "docs/governance/creative/index",
+                  "docs/governance/creative/get_creative_features"
+                ]
               }
             ]
           },

docs/governance/creative/get_creative_features.mdx

@@ -0,0 +1,172 @@
+---
+title: get_creative_features
+---
+
+# get_creative_features
+
+<Info>
+**AdCP 3.0 Proposal** - This task is under development for AdCP 3.0.
+</Info>
+
+Evaluates a creative manifest and returns feature values from a creative governance agent. The creative analog of `get_property_features`.
+
+## Use cases
+
+- **Security scanning**: Detect malware, auto-redirects, credential harvesting, cloaking
+- **Creative quality**: Evaluate brand consistency, platform optimization, guideline adherence
+- **Content categorization**: Classify creative content against IAB Content Taxonomy or other standards
+- **Accessibility**: Check WCAG compliance, screen reader compatibility
+
+## Request
+
+```json
+{
+  "$schema": "/schemas/creative/get-creative-features-request.json",
+  "creative_manifest": {
+    "format_id": {
+      "agent_url": "https://creative.adcontextprotocol.org",
+      "id": "html5-display-300x250"
+    },
+    "assets": {
+      "creative_html": {
+        "url": "https://cdn.agency.com/creative/abc123.html"
+      }
+    }
+  },
+  "feature_ids": ["auto_redirect", "credential_harvest", "cloaking"]
+}
+```
+
+### Parameters
+
+| Parameter | Type | Required | Description |
+|-----------|------|----------|-------------|
+| `creative_manifest` | object | Yes | Creative manifest with `format_id` and `assets` |
+| `feature_ids` | string[] | No | Filter to specific features. If omitted, evaluates all features the agent supports. |
+
+## Response
+
+<CodeGroup>
+```json Security scanner (clean)
+{
+  "$schema": "/schemas/creative/get-creative-features-response.json",
+  "results": [
+    { "feature_id": "auto_redirect", "value": false },
+    { "feature_id": "credential_harvest", "value": false },
+    { "feature_id": "cloaking", "value": false }
+  ],
+  "detail_url": "https://scanner.example.com/reports/ctx_abc123"
+}
+```
+
+```json Security scanner (threat detected)
+{
+  "$schema": "/schemas/creative/get-creative-features-response.json",
+  "results": [
+    { "feature_id": "auto_redirect", "value": true, "confidence": 0.97 },
+    { "feature_id": "credential_harvest", "value": true, "confidence": 0.91 },
+    { "feature_id": "cloaking", "value": false }
+  ],
+  "detail_url": "https://scanner.example.com/reports/ctx_def456"
+}
+```
+
+```json Creative quality platform
+{
+  "$schema": "/schemas/creative/get-creative-features-response.json",
+  "results": [
+    { "feature_id": "brand_consistency", "value": 87, "unit": "percentage" },
+    { "feature_id": "platform_optimized", "value": true },
+    { "feature_id": "creative_quality_score", "value": 92, "unit": "score" }
+  ],
+  "detail_url": "https://quality.example.com/reports/ctx_ghi789"
+}
+```
+
+```json Content categorizer
+{
+  "$schema": "/schemas/creative/get-creative-features-response.json",
+  "results": [
+    { "feature_id": "iab_casinos_gambling", "value": true, "confidence": 0.95 },
+    { "feature_id": "iab_automotive", "value": false, "confidence": 0.12 }
+  ],
+  "detail_url": "https://categorizer.example.com/reports/ctx_jkl012"
+}
+```
+</CodeGroup>
+
+### Response fields
+
+| Field | Description |
+|-------|-------------|
+| `results` | Array of feature evaluation results |
+| `results[].feature_id` | Which feature was evaluated |
+| `results[].value` | Feature value: boolean (binary), number (quantitative), or string (categorical) |
+| `results[].confidence` | Confidence score (0-1), when applicable |
+| `results[].unit` | Unit for quantitative values (e.g., `percentage`, `score`) |
+| `results[].expires_at` | When this evaluation expires and should be refreshed |
+| `results[].measured_at` | When this feature was evaluated |
+| `results[].methodology_version` | Version of methodology used |
+| `results[].details` | Vendor-specific details |
+| `detail_url` | URL to vendor's full assessment. Access-controlled by the vendor. |
+
+### Error response
+
+```json
+{
+  "errors": [
+    {
+      "code": "CREATIVE_INACCESSIBLE",
+      "message": "Could not retrieve creative assets for evaluation"
+    }
+  ]
+}
+```
+
+## Async evaluation
+
+Some evaluations (e.g., sandboxed malware scanning) take time. The agent returns `status: "working"` and delivers results via webhook when complete. This uses the standard [async task pattern](/docs/building/implementation/async-operations) — no custom status values needed.
+
+## Orchestrator logic
+
+The orchestrator applies feature requirements on the client side, the same way property list feature requirements work:
+
+```javascript
+const result = await agent.getCreativeFeatures({
+  creative_manifest: manifest
+});
+
+if (result.errors) {
+  // Handle error - reject or retry
+  return;
+}
+
+// Apply security requirements
+const threats = result.results.filter(
+  f => ['auto_redirect', 'credential_harvest', 'cloaking'].includes(f.feature_id)
+    && f.value === true
+);
+if (threats.length > 0) {
+  // Reject - security threat detected
+  return;
+}
+
+// Apply quality requirements
+const quality = result.results.find(f => f.feature_id === 'brand_consistency');
+if (quality && quality.value < 80) {
+  // Reject - below quality threshold
+  return;
+}
+```
+
+## Relationship to property governance
+
+Creative governance follows the same pattern as property governance:
+
+| Concept | Property governance | Creative governance |
+|---------|-------------------|-------------------|
+| **What's evaluated** | Properties (websites, apps) | Creatives (manifests) |
+| **Feature declarations** | `governance.property_features` | `governance.creative_features` |
+| **Evaluation task** | `get_property_features` | `get_creative_features` |
+| **Feature values** | `property-feature-value` schema | Same fields (value, confidence, expires_at, etc.) |
+| **Detailed intelligence** | Behind `detail_url` / `methodology_url` | Behind `detail_url` / `methodology_url` |

docs/governance/creative/index.mdx

@@ -0,0 +1,157 @@
+---
+title: Creative Governance
+sidebarTitle: Overview
+---
+
+<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>
+
+Creative Governance standardizes how creatives are evaluated by specialized governance agents. It applies the same feature-based pattern as [Property Governance](../property/index) — agents declare features they can evaluate, accept creative manifests, and return feature values.
+
+## Overview
+
+Creative governance agents evaluate creatives and return feature values. Different agents evaluate different features:
+
+| Agent type | Example features | Feature type |
+|------------|-----------------|--------------|
+| **Security scanner** | `auto_redirect`, `credential_harvest`, `cloaking` | Binary |
+| **Creative quality** | `brand_consistency`, `platform_optimized`, `creative_quality_score` | Quantitative, binary |
+| **Content categorization** | `iab_casinos_gambling`, `iab_automotive` | Binary (with confidence) |
+
+The protocol doesn't define a fixed feature taxonomy. Vendors declare what they evaluate via [`get_adcp_capabilities`](/docs/protocol/get_adcp_capabilities) and compete on coverage.
+
+## How it works
+
+### 1. Agent declares features
+
+A creative governance agent advertises its capabilities using the same feature definition pattern as property governance:
+
+<CodeGroup>
+```json Security scanner
+{
+  "governance": {
+    "creative_features": [
+      {
+        "feature_id": "auto_redirect",
+        "type": "binary",
+        "description": "Unauthorized navigation away from publisher context without user interaction",
+        "methodology_url": "https://scanner.example.com/methodology"
+      },
+      {
+        "feature_id": "credential_harvest",
+        "type": "binary",
+        "description": "Phishing techniques to gather user credentials or PII",
+        "methodology_url": "https://scanner.example.com/methodology"
+      },
+      {
+        "feature_id": "cloaking",
+        "type": "binary",
+        "description": "Creative masks or misrepresents content to evade detection",
+        "methodology_url": "https://scanner.example.com/methodology"
+      }
+    ]
+  }
+}
+```
+
+```json Creative quality platform
+{
+  "governance": {
+    "creative_features": [
+      {
+        "feature_id": "brand_consistency",
+        "type": "quantitative",
+        "range": { "min": 0, "max": 100 },
+        "description": "Adherence to brand guidelines including logo placement, colors, and typography",
+        "methodology_url": "https://quality.example.com/methodology"
+      },
+      {
+        "feature_id": "platform_optimized",
+        "type": "binary",
+        "description": "Creative meets platform-specific best practices (aspect ratio, text overlay limits)",
+        "methodology_url": "https://quality.example.com/methodology"
+      }
+    ]
+  }
+}
+```
+
+```json Content categorizer
+{
+  "governance": {
+    "creative_features": [
+      {
+        "feature_id": "iab_casinos_gambling",
+        "type": "binary",
+        "description": "Creative contains casinos or gambling content (IAB Content Taxonomy 3.1, ID 181)",
+        "methodology_url": "https://categorizer.example.com/methodology"
+      },
+      {
+        "feature_id": "iab_automotive",
+        "type": "binary",
+        "description": "Creative contains automotive content (IAB Content Taxonomy 3.1, ID 1)",
+        "methodology_url": "https://categorizer.example.com/methodology"
+      }
+    ]
+  }
+}
+```
+</CodeGroup>
+
+### 2. Orchestrator evaluates a creative
+
+The orchestrator calls [`get_creative_features`](./get_creative_features) with a creative manifest:
+
+```json
+{
+  "creative_manifest": {
+    "format_id": {
+      "agent_url": "https://creative.adcontextprotocol.org",
+      "id": "html5-display-300x250"
+    },
+    "assets": {
+      "creative_html": {
+        "url": "https://cdn.agency.com/creative/abc123.html"
+      }
+    }
+  }
+}
+```
+
+### 3. Agent returns feature values
+
+The agent evaluates the creative and returns feature values. The response shape is the same regardless of agent type:
+
+```json
+{
+  "results": [
+    { "feature_id": "auto_redirect", "value": true, "confidence": 0.97 },
+    { "feature_id": "credential_harvest", "value": false },
+    { "feature_id": "cloaking", "value": false }
+  ],
+  "detail_url": "https://scanner.example.com/reports/ctx_abc123"
+}
+```
+
+### 4. Orchestrator applies requirements
+
+The orchestrator evaluates the feature values against buyer-defined requirements — the same pattern as property list feature requirements:
+
+- Security: reject if `auto_redirect` is `true`
+- Quality: reject if `brand_consistency` is below 80
+- Categorization: reject if `iab_casinos_gambling` is `true` and campaign excludes gambling
+
+## Design principles
+
+**Reuses existing patterns.** Creative governance follows the same feature-based model as property governance. Agents declare features, return values, and keep detailed intelligence behind `detail_url` and `methodology_url`.
+
+**Vendor-agnostic features.** The protocol doesn't define which features exist. Vendors declare their own feature catalogs. A security vendor and a creative quality vendor both use `get_creative_features` — the protocol doesn't need to know the difference.
+
+**Opaque results.** Feature values on the wire are pass/fail (binary) or scores (quantitative). Detailed methodology, detection techniques, and threat intelligence stay behind the vendor's access-controlled URLs.
+
+**Orchestrator decides.** Agents return values, orchestrators evaluate against requirements. The separation of measurement from decision-making is the same as property governance.
+
+## Async evaluation
+
+If evaluation takes time (e.g., sandboxed execution for malware scanning), the agent returns `status: "working"` and delivers results via the standard webhook mechanism. No custom status values or webhook events needed — the existing [async task pattern](/docs/building/implementation/async-operations) handles this.

docs/governance/overview.mdx

@@ -22,7 +22,7 @@ The Governance Protocol covers four distinct governance domains:
 | **[Property Governance](./property/index)** | Where ads can run | Property lists, compliance filtering, adagents.json authorization |
 | **[Brand Protocol](/docs/brand-protocol/index)** | Brand identity & agents | brand.json discovery, brand agent authorization |
 | **[Content Standards](./content-standards/index)** | Brand suitability | Privacy-preserving content evaluation, calibration, validation |
-| **Creative Governance** | What creatives are compliant | Format validation, content moderation, accessibility |
+| **[Creative Governance](./creative/index)** | What creatives are safe and effective | Security scanning, creative quality, content categorization via `get_creative_features` |
 | **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.