Documentation Index
Fetch the complete documentation index at: https://docs.adcontextprotocol.org/llms.txt
Use this file to discover all available pages before exploring further.
Discover available advertising products based on campaign requirements using natural language briefs or structured filters.
Why this shape. Targeting, pricing, and curation are folded into one round-trip — the brief drives discovery, the publisher curates against it, and pricing_options carry firm prices the buyer commits against via pricing_option_id. We rejected a separate get_price_quote step between products and buy creation: it splits one expert decision into two underspecified ones and breaks the brief→curation contract. Iteration is buying_mode: "refine" with a typed change array — not a new task. → Design principle: the brief drives discovery.
Authentication: Optional (returns limited results without credentials)
Response Time: ~60 seconds (AI inference with back-end systems)
Request Schema: /schemas/v3/media-buy/get-products-request.json
Response Schema: /schemas/v3/media-buy/get-products-response.json
Quick Start
Discover products with a natural language brief:
import { testAgent } from '@adcp/sdk/testing';
import { GetProductsResponseSchema } from '@adcp/sdk';
const result = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'Premium athletic footwear with innovative cushioning',
brand: {
domain: 'acmecorp.com'
}
});
if (!result.success) {
throw new Error(`Request failed: ${result.error}`);
}
// Validate response against schema
const validated = GetProductsResponseSchema.parse(result.data);
console.log(`Found ${validated.products.length} products`);
// Access validated product fields
for (const product of validated.products) {
console.log(`- ${product.name} (${product.delivery_type})`);
console.log(` Formats: ${product.format_ids.map(f => f.id).join(', ')}`);
}
Using Structured Filters
You can also use structured filters instead of (or in addition to) a brief. In brief mode, filters act as hard constraints on top of the publisher’s curation — the brief describes intent, filters enforce requirements:
import { testAgent } from '@adcp/sdk/testing';
const result = await testAgent.getProducts({
buying_mode: 'wholesale',
brand: {
domain: 'acmecorp.com'
},
filters: {
channels: ['ctv'],
delivery_type: 'guaranteed',
standard_formats_only: true
}
});
if (result.success && result.data) {
console.log(`Found ${result.data.products.length} guaranteed CTV products`);
}
Request Parameters
| Parameter | Type | Required | Description |
|---|
buying_mode | string | Yes | "brief", "wholesale", or "refine". "brief": publisher curates products from the brief. "wholesale": raw inventory for buyer-directed targeting, brief must not be provided. "refine": iterate on products and proposals from a previous response using the refine array of change requests. v3 clients MUST include buying_mode. Sellers receiving requests from pre-v3 clients without buying_mode SHOULD default to "brief". Timing semantics: "wholesale" is a catalog read — sellers SHOULD return a synchronous response and MUST NOT route a "wholesale" request through the async/Submitted arm. Partial completion is signalled via incomplete[], not a task handoff. "brief" and "refine" MAY complete synchronously OR MAY return a Submitted envelope when curation requires upstream-system queries or HITL review the seller cannot complete inside time_budget. Buyers needing predictable fast catalog access MUST use "wholesale". |
brief | string | Conditional | Natural language description of campaign requirements. Required when buying_mode is "brief". Must not be provided when buying_mode is "wholesale" or "refine". |
refine | Refine[] | Conditional | Array of change requests for iterating on products and proposals. Required when buying_mode is "refine". Must not be provided when buying_mode is "brief" or "wholesale". See Refine array below. |
brand | BrandRef | No | Brand reference (domain + optional brand_id). Resolved to full identity at execution time. |
account | AccountRef | No | Account reference for account-specific pricing. Returns products with pricing from this account’s rate card. |
catalog | Catalog | No | Catalog of items the buyer wants to promote. The seller matches catalog items against its inventory and returns products where matches exist. Requires brand. See Catalog discovery below. |
filters | Filters | No | Structured filters (see below) |
property_list | PropertyListRef | No | [AdCP 3.0] Reference to a property list for filtering. See Property Lists |
pagination | PaginationRequest | No | Cursor-based pagination for large product catalogs (see below) |
time_budget | Duration | No | Maximum time the buyer will commit to this request. The seller returns the best results achievable within this budget and does not start processes (human approvals, expensive external queries) that cannot complete in time. When omitted, the seller decides timing. Example: {"interval": 30, "unit": "seconds"}. |
Property GovernanceThe property_list filter references a property list created via create_property_list on a property governance agent. Property lists define which publisher properties meet compliance requirements — COPPA-certified sites, sustainability-scored inventory, brand-safe publishers, etc.To use property list filtering:
- Call
get_adcp_capabilities on a property governance agent to discover available property_features
- Create a property list via
create_property_list with your feature requirements
- Pass the resulting
property_list_id to get_products to filter inventory
The seller must declare features.property_list_filtering: true in get_adcp_capabilities to support this filter. See the Property Governance overview for the full workflow.
Filters Object
| Parameter | Type | Description |
|---|
delivery_type | string | Filter by "guaranteed" or "non_guaranteed" |
is_fixed_price | boolean | Filter for fixed price vs auction products |
format_ids | FormatID[] | Filter by specific format IDs |
standard_formats_only | boolean | Only return products accepting IAB standard formats |
min_exposures | integer | Minimum exposures needed for measurement validity |
start_date | string | Campaign start date in ISO 8601 format (YYYY-MM-DD) for availability checks |
end_date | string | Campaign end date in ISO 8601 format (YYYY-MM-DD) for availability checks |
budget_range | object | Budget range to filter appropriate products (see Budget Range Object below) |
countries | string[] | Filter by target countries using ISO 3166-1 alpha-2 codes (e.g., ["US", "CA", "GB"]) |
regions | string[] | Filter by region coverage using ISO 3166-2 codes (e.g., ["US-NY", "GB-SCT"]). Best for locally-bound inventory |
metros | object[] | Filter by metro coverage. Each entry: { system, code } (e.g., [{ "system": "nielsen_dma", "code": "501" }]) |
channels | string[] | Filter by advertising channels (e.g., ["display", "ctv", "social", "streaming_audio"]). See Media Channel Taxonomy |
postal_areas | object[] | Filter by postal area coverage. Each entry: { system, values } (e.g., [{ "system": "us_zip", "values": ["10001"] }]) |
geo_proximity | object[] | Filter by proximity to geographic points. Each entry uses exactly one boundary method: radius, travel_time + transport_mode, or geometry |
keywords | object[] | Filter by keyword relevance for search/retail media. Each entry: { keyword, match_type? }. match_type defaults to broad if omitted |
required_performance_standards | PerformanceStandard[] | Filter to products that can meet the buyer’s performance standard requirements. Each entry specifies a metric, threshold, and vendor (e.g., “DoubleVerify for viewability at 70% MRC”). Products that cannot meet these thresholds or do not support the specified vendors are excluded. |
required_metrics | string[] (metric vocabulary) | Filter to products whose reporting_capabilities.available_metrics is a superset of these metrics — i.e., products that commit to reporting all listed metrics in delivery. Use for capability discovery (e.g., ["completed_views"] for a CTV CPCV buy). Sellers MUST silently exclude products that cannot meet the list — filter-not-fail; do not return an error. The product’s declared available_metrics becomes the binding reporting contract carried into the resulting media buy. |
required_vendor_metrics | object[] | Filter to products whose reporting_capabilities.vendor_metrics covers vendor-defined metrics (proprietary attention, emissions, panel demographics, brand-lift surveys, etc.). Each entry pins vendor (BrandRef) and/or metric_id — at least one. Cross-vendor discovery (e.g., “any attention measurement”) is the buyer agent’s responsibility: resolve which vendors offer a category via the vendors’ brand.json records, then enumerate them as filter entries. Same filter-not-fail semantics as required_metrics. |
Budget Range Object
| Parameter | Type | Required | Description |
|---|
currency | string | Yes | ISO 4217 currency code (e.g., "USD", "EUR", "GBP") |
min | number | No* | Minimum budget amount |
max | number | No* | Maximum budget amount |
*At least one of min or max must be specified.
Refine array
The refine array is a list of change requests. Each entry declares a scope and what the buyer is asking for. At least one entry is required. The seller considers all entries together when composing the response, and replies to each via refinement_applied.
Each entry is a discriminated union on scope:
scope: “request”
| Field | Type | Required | Description |
|---|
scope | string | Yes | "request" |
ask | string | Yes | Direction for the selection as a whole (e.g., "more video options", "suggest how to combine these products"). |
scope: “product”
| Field | Type | Required | Description |
|---|
scope | string | Yes | "product" |
product_id | string | Yes | Product ID from a previous get_products response |
action | string | No | "include" (default): return this product with updated pricing and data. "omit": exclude from the response. "more_like_this": find similar products (the original is also returned). When omitted, the seller treats the entry as "include". |
ask | string | No | What the buyer is asking for. For "include": specific changes (e.g., "add 16:9 format"). For "more_like_this": what “similar” means (e.g., "same audience but video format"). Ignored when action is "omit". |
scope: “proposal”
| Field | Type | Required | Description |
|---|
scope | string | Yes | "proposal" |
proposal_id | string | Yes | Proposal ID from a previous get_products response |
action | string | No | "include" (default): return with updated allocations and pricing. "omit": exclude from the response. "finalize": request firm pricing and inventory hold (transitions a draft proposal to committed). When omitted, the seller treats the entry as "include". |
ask | string | No | What the buyer is asking for (e.g., "shift more budget toward video", "reduce total by 10%"). Ignored when action is "omit". |
refinement_applied (response)
When the seller receives a refine array, the response includes refinement_applied — an array matched by position. Each entry reports whether the ask was fulfilled:
| Field | Type | Required | Description |
|---|
scope | string | Yes | Echoes the scope ("request" / "product" / "proposal") from the corresponding refine entry. |
product_id | string | Yes when scope is "product" | Echoes product_id from the corresponding refine entry. |
proposal_id | string | Yes when scope is "proposal" | Echoes proposal_id from the corresponding refine entry. |
status | string | Yes | "applied": ask fulfilled. "partial": partially fulfilled. "unable": could not fulfill. |
notes | string | No | Seller explanation. Recommended when status is "partial" or "unable". |
Catalog discovery
Pass a catalog to find advertising products that can promote your catalog items. The seller matches your catalog items against its inventory and returns products where matches exist. Supports all catalog types — a product catalog finds sponsored product slots, a job catalog finds job ad products, a flight catalog finds dynamic travel ads.
The catalog field uses the same Catalog object used throughout AdCP. You can reference a synced catalog by catalog_id, provide inline items, or use selectors to filter:
| Field | Type | Description |
|---|
type | CatalogType | Catalog type (required) — product, job, hotel, flight, offering, etc. |
catalog_id | string | Reference a synced catalog by ID |
ids | string[] | Filter to specific item IDs |
gtins | string[] | Filter by GTIN for cross-retailer matching (product type only) |
tags | string[] | Filter by tags (OR logic) |
category | string | Filter by category |
query | string | Natural language filter |
Products in the response include catalog_types (what catalog types they support) and catalog_match (which items matched).
Response
Returns an array of products and optionally proposals.
Products Array
| Field | Type | Description |
|---|
product_id | string | Unique product identifier |
name | string | Human-readable product name |
description | string | Detailed product description |
publisher_properties | PublisherProperty[] | Array of publisher entries, each with publisher_domain and either property_ids or property_tags |
format_ids | FormatID[] | Supported creative format IDs |
delivery_type | string | "guaranteed" or "non_guaranteed" |
delivery_measurement | DeliveryMeasurement | (Optional) How delivery is measured (impressions, views, etc.) |
pricing_options | PricingOption[] | Available pricing models (CPM, CPCV, etc.). Auction options may include floor_price and optional price_guidance. Bid-based auction models (CPM, vCPM, CPC, CPCV, CPV) may also include optional max_bid (boolean). |
shows | CollectionSelector[] | (Optional) Collections available in this product. Each entry has publisher_domain and collection_ids. Buyers resolve full collection objects from the referenced adagents.json. See Collections and installments. |
collection_targeting_allowed | boolean | (Optional, default: false) Whether buyers can target a subset of this product’s shows. When false, the product is a bundle. |
brief_relevance | string | Why this product matches the brief (when brief provided) |
measurement_readiness | MeasurementReadiness | (Optional) Whether the buyer’s event setup is sufficient for this product’s optimization. Only present when the seller can evaluate the buyer’s account context. |
measurement_terms | MeasurementTerms | (Optional) Seller’s default billing measurement and makegood terms. Buyers may propose different terms at create_media_buy. |
performance_standards | PerformanceStandard[] | (Optional) Seller’s default performance standards (viewability, IVT, completion rate, brand safety, attention score). Buyers may propose different standards at create_media_buy. |
cancellation_policy | CancellationPolicy | (Optional) Cancellation notice period and penalties for guaranteed products. Buyers accept these terms by creating a media buy against the product. |
Proposals Array (Optional)
Publishers may return proposals alongside products - structured media plans with budget allocations. See Proposals for details.
| Field | Type | Description |
|---|
proposal_id | string | Unique identifier for executing this proposal via create_media_buy |
name | string | Human-readable name for the media plan |
allocations | ProductAllocation[] | Budget allocations across products (percentages must sum to 100). Each allocation may include optional start_time and end_time for per-flight scheduling. |
forecast | DeliveryForecast | Aggregate delivery forecast for the proposal. Contains forecast points with metric ranges. See Delivery Forecasts |
total_budget_guidance | object | Optional min/recommended/max budget guidance |
brief_alignment | string | How this proposal addresses the campaign brief |
expires_at | string | ISO 8601 timestamp when this proposal expires |
For large product catalogs, use cursor-based pagination:
| Request Parameter | Type | Description |
|---|
pagination.max_results | integer | Maximum products per page (1-100, default: 50) |
pagination.cursor | string | Cursor from previous response for next page |
| Response Field | Type | Description |
|---|
pagination.has_more | boolean | Whether more products are available |
pagination.cursor | string | Cursor to pass for the next page |
pagination.total_count | integer | Total matching products (optional, not all backends support this) |
Pagination is optional. When omitted, the server returns all results (or a server-chosen default page). When the response includes pagination.has_more: true, pass pagination.cursor in the next request to get the next page.
| Field | Type | Description |
|---|
property_list_applied | boolean | [AdCP 3.0] true if the agent filtered products based on the provided property_list. Absent or false if not provided or not supported. |
catalog_applied | boolean | true if the seller filtered results based on the provided catalog. Absent or false if no catalog was provided or the seller does not support catalog matching. |
refinement_applied | RefinementResult[] | Seller acknowledgment of each refine entry, matched by position. Only present when buying_mode is "refine". See refinement_applied above. |
incomplete | IncompleteEntry[] | Declares what the seller could not finish within the time_budget or due to internal limits. Each entry identifies a scope with a human-readable explanation. Absent when the response is fully complete. See incomplete array below. |
filter_diagnostics | object | Optional non-fatal observability block describing how filters narrowed the candidate set — total_candidates plus per-filter excluded_by counts (keyed by filter name). Disambiguates “no inventory” from “your filter excluded everything” when the result list is empty or unexpectedly small. Counts only — never product names — to avoid leaking competitive intelligence. See filter_diagnostics below. |
filter_diagnostics
When the seller can attribute exclusions to specific filters, the response MAY include a filter_diagnostics block. This is observability — not error reporting; sellers still silently exclude unmatched products per the filter-not-fail convention. Buyers use this to triage empty/small results without depending on its presence. total_candidates and excluded_by are independently optional — sellers whose baseline catalog size is sensitive MAY emit excluded_by without total_candidates.
| Field | Type | Description |
|---|
semantics | string | "only" (deterministic; counts products that would have been included if not for this filter alone — recommended for triage), "any" (counts products excluded by any filter; counts may overlap), or "approximate" (seller can’t cleanly attribute exclusions to a single filter). Buyers SHOULD inspect semantics before doing arithmetic on counts. |
total_candidates | integer | Number of products considered before filters were applied. May be sampled or capped at large catalogs. Optional. |
excluded_by | object | Keys are filter property names from the request (required_metrics, required_geo_targeting, budget_range, etc.). Each value is { count, values?, notes? }. Only filters that meaningfully narrowed the set need appear. |
excluded_by.<filter>.count | integer | Count of products excluded by this filter, interpreted per the parent semantics field. |
excluded_by.<filter>.values | array | Optional list of the specific filter values that contributed to exclusions (e.g., ["completed_views"] for required_metrics). Items are strings or objects depending on filter shape; opaque without filter-specific knowledge. |
excluded_by.<filter>.notes | string | Optional human-readable note about the narrowing. |
{
"products": [],
"filter_diagnostics": {
"semantics": "only",
"total_candidates": 47,
"excluded_by": {
"required_metrics": { "count": 31, "values": ["completed_views"] },
"required_geo_targeting": { "count": 9 },
"budget_range": { "count": 7 }
}
}
}
incomplete array
When the seller cannot complete all work within the time_budget (or due to its own internal limits), the response includes incomplete — an array declaring what is missing. Buyers can use estimated_wait to decide whether to retry with a larger budget.
| Field | Type | Required | Description |
|---|
scope | string | Yes | "products": not all inventory sources were searched. "pricing": products returned but pricing is absent or unconfirmed. "forecast": products returned but forecast data is absent. "proposals": proposals were not generated or are incomplete. |
description | string | Yes | Human-readable explanation of what is missing and why. |
estimated_wait | Duration | No | How much additional time would resolve this scope. |
See schema for complete field list: get-products-response.json
Common Scenarios
Time-budgeted discovery
Declare a time budget when you need fast results and can accept partial data. The seller returns what it can within the budget and declares what is incomplete:
import { testAgent } from '@adcp/sdk/testing';
const result = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'CTV and display for brand awareness',
brand: {
domain: 'acmecorp.com'
},
time_budget: {
interval: 10,
unit: 'seconds'
}
});
if (result.success && result.data) {
console.log(`Found ${result.data.products.length} products`);
if (result.data.incomplete) {
for (const entry of result.data.incomplete) {
console.log(`Incomplete: ${entry.scope} — ${entry.description}`);
if (entry.estimated_wait) {
console.log(` Would resolve in ${entry.estimated_wait.interval} ${entry.estimated_wait.unit}`);
}
}
}
}
A response with incomplete data — products are returned but some scopes are missing:
{
"products": [
{
"product_id": "prog-display-ros",
"name": "Programmatic Display — Run of Site",
"delivery_type": "non_guaranteed",
"pricing_options": [{ "pricing_option_id": "cpm-ros", "pricing_model": "cpm", "currency": "USD", "fixed_price": 12.00 }]
}
],
"incomplete": [
{
"scope": "products",
"description": "Premium inventory not searched — requires publisher approval",
"estimated_wait": { "interval": 60, "unit": "minutes" }
},
{
"scope": "forecast",
"description": "Forecast model did not complete within budget",
"estimated_wait": { "interval": 45, "unit": "seconds" }
}
]
}
Standard Catalog Discovery
import { testAgent } from '@adcp/sdk/testing';
// wholesale mode: buyer applies their own audiences, no publisher curation
const result = await testAgent.getProducts({
buying_mode: 'wholesale',
brand: {
domain: 'acmecorp.com'
},
filters: {
delivery_type: 'non_guaranteed'
}
});
if (result.success && result.data) {
console.log(`Found ${result.data.products.length} standard catalog products`);
}
import { testAgent } from '@adcp/sdk/testing';
// Find products supporting both video and display
const result = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'Brand awareness campaign with video and display',
brand: {
domain: 'acmecorp.com'
},
filters: {
channels: ['display', 'ctv']
}
});
if (result.success && result.data) {
console.log(`Found ${result.data.products.length} products supporting video and display`);
}
Budget and Date Filtering
import { testAgent } from '@adcp/sdk/testing';
// Find products within budget and date range for specific countries and channels
const result = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'Q2 campaign for athletic footwear in North America',
brand: {
domain: 'acmecorp.com'
},
filters: {
start_date: '2025-04-01',
end_date: '2025-06-30',
budget_range: {
min: 50000,
max: 100000,
currency: 'USD'
},
countries: ['US', 'CA'],
channels: ['display', 'ctv', 'podcast'],
delivery_type: 'guaranteed'
}
});
if (result.success && result.data) {
console.log(`Found ${result.data.products.length} products for Q2 within budget`);
}
Property Tag Resolution
import { testAgent } from '@adcp/sdk/testing';
// Get products with property tags
const result = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'Sports content',
brand: {
domain: 'acmecorp.com'
}
});
if (result.success && result.data) {
// Products with property_tags in publisher_properties represent large networks
// Use get_adcp_capabilities to discover the agent's portfolio
const productsWithTags = result.data.products.filter(p =>
p.publisher_properties?.some(pub => pub.property_tags && pub.property_tags.length > 0)
);
console.log(`${productsWithTags.length} products use property tags (large networks)`);
}
Guaranteed Delivery Products
import { testAgent } from '@adcp/sdk/testing';
// Find guaranteed delivery products for measurement
const result = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'Guaranteed delivery for lift study',
brand: {
domain: 'acmecorp.com'
},
filters: {
delivery_type: 'guaranteed',
min_exposures: 100000
}
});
if (result.success && result.data) {
console.log(`Found ${result.data.products.length} guaranteed products with 100k+ exposures`);
}
import { testAgent } from '@adcp/sdk/testing';
// Find products that only accept IAB standard formats
const result = await testAgent.getProducts({
buying_mode: 'wholesale',
brand: {
domain: 'acmecorp.com'
},
filters: {
standard_formats_only: true
}
});
if (result.success && result.data) {
console.log(`Found ${result.data.products.length} products with standard formats only`);
}
Catalog-driven discovery
Use catalog with a brand to discover advertising products that can promote your catalog items. The seller matches your items against its inventory and returns products where matches exist:
import { testAgent } from '@adcp/sdk/testing';
// Discover retail media products for specific catalog items
const result = await testAgent.getProducts({
buying_mode: 'wholesale',
brand: {
domain: 'acmecorp.com'
},
catalog: {
type: 'product',
tags: ['ketchup', 'organic'],
category: 'food/condiments'
},
filters: {
channels: ['retail_media']
}
});
if (result.success && result.data) {
if (result.data.catalog_applied) {
console.log(`Found ${result.data.products.length} products with catalog matches`);
} else {
console.log('Seller does not support catalog matching');
}
}
You can also use GTIN matching, reference a synced catalog, or discover products for other catalog types:
{
"catalog": {
"type": "product",
"gtins": ["00013000006040", "00013000006057"]
}
}
{
"catalog": {
"catalog_id": "gmc-primary",
"type": "product"
}
}
{
"catalog": {
"type": "job",
"catalog_id": "chef-vacancies"
}
}
Property List Filtering
AdCP 3.0 - Property list filtering requires governance agent support.
Filter products to only those available on properties in your approved list:
import { testAgent } from '@adcp/sdk/testing';
// Filter products by property list from governance agent
const result = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'Brand-safe inventory for family brand',
brand: {
domain: 'acmecorp.com'
},
property_list: {
agent_url: 'https://governance.example.com',
list_id: 'pl_brand_safe_2024'
}
});
if (result.success && result.data) {
// Check if filtering was actually applied
if (result.data.property_list_applied) {
console.log(`Found ${result.data.products.length} products on approved properties`);
} else {
console.log('Agent does not support property list filtering');
console.log(`Found ${result.data.products.length} products (unfiltered)`);
}
}
Note: If property_list_applied is absent or false, the sales agent did not filter products. This can happen if:
- The agent doesn’t support property governance features
- The agent couldn’t access the property list
- The property list had no effect on the available inventory
Property Targeting Behavior
Products have a property_targeting_allowed flag that affects filtering:
property_targeting_allowed: false (default): Product is “all or nothing” - excluded unless your list contains all of its properties
property_targeting_allowed: true: Product is included if there’s any intersection between its properties and your list
This allows publishers to offer run-of-network products that can’t be cherry-picked alongside flexible inventory that buyers can filter.
See Property Targeting for more details and Property Governance for more on property lists.
Refinement
After initial discovery, use buying_mode: "refine" to iterate on specific products and proposals. The refine array is a list of change requests — each entry declares a scope and what the buyer is asking for. The seller returns updated products with revised pricing and configurations, plus refinement_applied acknowledging each ask.
See the Refinement guide for the full walkthrough: scope types, action semantics, seller responses, and common patterns. The parameter shape is defined in the Refine array section above.
Minimal example:
{
"buying_mode": "refine",
"refine": [
{ "scope": "request", "ask": "more video, less display" },
{ "scope": "product", "product_id": "prod_premium_video", "ask": "add 16:9 format option" },
{ "scope": "product", "product_id": "prod_display_run_of_site", "action": "omit" },
{ "scope": "proposal", "proposal_id": "prop_awareness_q2", "ask": "reallocate display budget to video" }
],
"filters": {
"start_date": "2026-04-01",
"end_date": "2026-04-30",
"budget_range": { "min": 200000, "max": 200000, "currency": "USD" }
}
}
Key rules to know before sending:
refine is only valid in refine mode. Requests that include this field in brief or wholesale mode are rejected with INVALID_REQUEST.
- Filters are absolute, not deltas. Always send the full filter set you want applied.
- Proposals are ephemeral. Proposals typically include an
expires_at timestamp. After expiration, the seller returns PROPOSAL_EXPIRED.
- Product IDs are stable catalog identifiers. Custom products (
is_custom: true) may have an expires_at timestamp, after which refinement returns PRODUCT_NOT_FOUND.
Error Handling
| Error Code | Description | Resolution |
|---|
AUTH_MISSING | No credentials presented | Provide credentials via auth header |
AUTH_INVALID | Credentials rejected (expired / revoked) | Human credential rotation required; do not auto-retry |
INVALID_REQUEST | Brief too long or malformed filters | Check request parameters |
PRODUCT_NOT_FOUND | One or more referenced product IDs are unknown or expired | Remove invalid IDs and retry, or re-discover with a brief request |
PROPOSAL_EXPIRED | A referenced proposal ID has passed its expires_at timestamp | Re-discover with a new brief or wholesale request |
POLICY_VIOLATION | Category blocked for advertiser | See policy response message for details |
Authentication Comparison
See the difference between authenticated and unauthenticated access:
import { testAgent, testAgentNoAuth } from '@adcp/sdk/testing';
// WITH authentication - full catalog with pricing
const fullCatalog = await testAgent.getProducts({
buying_mode: 'brief',
brief: 'Premium CTV inventory for brand awareness',
brand: {
domain: 'acmecorp.com'
}
});
if (!fullCatalog.success) {
throw new Error(`Failed to get products: ${fullCatalog.error}`);
}
console.log(`With auth: ${fullCatalog.data.products.length} products`);
console.log(`First product pricing: ${fullCatalog.data.products[0].pricing_options.length} options`);
// WITHOUT authentication - limited public catalog
const publicCatalog = await testAgentNoAuth.getProducts({
buying_mode: 'brief',
brief: 'Premium CTV inventory for brand awareness',
brand: {
domain: 'acmecorp.com'
}
});
if (!publicCatalog.success) {
throw new Error(`Failed to get products: ${publicCatalog.error}`);
}
console.log(`Without auth: ${publicCatalog.data.products.length} products`);
console.log(`First product pricing: ${publicCatalog.data.products[0].pricing_options?.length || 0} options`);
Key Differences:
- Product Count: Authenticated access returns more products, including private/custom offerings
- Pricing Information: Only authenticated requests receive detailed pricing options (CPM, CPCV, etc.)
- Targeting Details: Custom targeting capabilities may be restricted to authenticated users
- Rate Limits: Unauthenticated requests have lower rate limits
Authentication Behavior
- Without credentials: Returns limited catalog (standard catalog products), no pricing, no custom offerings
- With credentials: Returns complete catalog with pricing and custom products
See Authentication Guide for details.
Asynchronous Operations
Most product searches complete immediately, but some scenarios require asynchronous processing. When this happens, you’ll receive a status other than completed and can track progress through webhooks or polling.
When Search Runs Asynchronously
Product search may require async processing in these situations:
- Complex searches: Searching across multiple inventory sources or custom curation
- Needs clarification: Your brief is vague and the system needs more information
- Custom products: Bespoke product packages that require human review
Async Status Flow
POST /api/mcp/call_tool
{
"name": "get_products",
"arguments": {
"buying_mode": "brief",
"brief": "CTV inventory for sports audience",
"brand": { "domain": "acmecorp.com" }
}
}
Response (200 OK):
{
"status": "completed",
"message": "Found 3 products matching your requirements",
"products": [...]
}
Needs Clarification
When the brief is unclear, the system asks for more details:Response (200 OK):
{
"status": "input-required",
"message": "I need a bit more information. What's your budget range and campaign duration?",
"task_id": "task_789",
"context_id": "ctx_123",
"reason": "CLARIFICATION_NEEDED",
"partial_results": [],
"suggestions": ["$50K-$100K", "1 month", "Q1 2024"]
}
Continue the conversation with the same context_id:POST /api/mcp/continue
{
"context_id": "ctx_123",
"message": "Budget is $75K for a 3-week campaign in March"
}
Response (200 OK):
{
"status": "completed",
"message": "Perfect! Found 5 products within your budget",
"products": [...]
}
Complex Search (With Webhook)
For searches requiring deep inventory analysis, configure a webhook:POST /api/mcp/call_tool
{
"name": "get_products",
"arguments": {
"buying_mode": "brief",
"brief": "Premium inventory across all formats for luxury automotive brand",
"brand": { "domain": "acmecorp.com" },
"pushNotificationConfig": {
"url": "https://buyer.com/webhooks/adcp/get_products",
"authentication": {
"schemes": ["Bearer"],
"credentials": "secret_token_32_chars"
}
}
}
}
Response (200 OK):
{
"status": "working",
"message": "Searching premium inventory across display, video, and audio",
"task_id": "task_456",
"context_id": "ctx_123",
"percentage": 10,
"current_step": "searching_inventory"
}
// Later, webhook POST to https://buyer.com/webhooks/adcp/get_products
{
"task_id": "task_456",
"task_type": "get_products",
"status": "completed",
"timestamp": "2025-01-22T10:30:00Z",
"message": "Found 12 premium products across all formats",
"result": {
"products": [...]
}
}
POST /api/a2a
{
"message": {
"role": "user",
"parts": [{
"kind": "data",
"data": {
"skill": "get_products",
"parameters": {
"buying_mode": "brief",
"brief": "CTV inventory for sports audience",
"brand": { "domain": "acmecorp.com" }
}
}
}]
}
}
Response (200 OK):
{
"id": "task_123",
"contextId": "ctx_456",
"artifact": {
"kind": "data",
"data": {
"products": [...]
}
},
"status": {
"state": "completed",
"message": {
"role": "agent",
"parts": [{ "text": "Found 3 products matching your requirements" }]
}
}
}
Needs Clarification
Real-time updates via SSE when clarification is needed:// Initial response
{
"id": "task_789",
"contextId": "ctx_123",
"status": {
"state": "input-required",
"message": {
"role": "agent",
"parts": [
{ "text": "I need a bit more information. What's your budget range and campaign duration?" },
{
"data": {
"reason": "CLARIFICATION_NEEDED",
"suggestions": ["$50K-$100K", "1 month", "Q1 2024"]
}
}
]
}
}
}
// Send follow-up
POST /api/a2a
{
"contextId": "ctx_123",
"message": {
"role": "user",
"parts": [{ "text": "Budget is $75K for a 3-week campaign in March" }]
}
}
// SSE update: task completed
{
"id": "task_789",
"contextId": "ctx_123",
"artifact": {
"kind": "data",
"data": { "products": [...] }
},
"status": {
"state": "completed",
"message": {
"role": "agent",
"parts": [{ "text": "Perfect! Found 5 products within your budget" }]
}
}
}
Complex Search (With Webhook)
Configure push notifications for long searches:POST /api/a2a
{
"message": {
"role": "user",
"parts": [{
"kind": "data",
"data": {
"skill": "get_products",
"parameters": {
"buying_mode": "brief",
"brief": "Premium inventory across all formats for luxury automotive brand",
"brand": { "domain": "acmecorp.com" }
}
}
}]
},
"pushNotificationConfig": {
"url": "https://buyer.com/webhooks/a2a/get_products",
"authentication": {
"schemes": ["bearer"],
"credentials": "secret_token_32_chars"
}
}
}
Response (200 OK):
{
"id": "task_456",
"contextId": "ctx_789",
"status": {
"state": "working",
"message": {
"role": "agent",
"parts": [
{ "text": "Searching premium inventory across display, video, and audio" },
{
"data": {
"percentage": 10,
"current_step": "searching_inventory"
}
}
]
}
}
}
// Later, webhook POST to https://buyer.com/webhooks/a2a/get_products
{
"id": "task_456",
"contextId": "ctx_789",
"artifact": {
"kind": "data",
"data": {
"products": [...]
}
},
"status": {
"state": "completed",
"message": {
"role": "agent",
"parts": [
{ "text": "Found 12 premium products across all formats" },
{
"data": {
"products": [...]
}
}
]
},
"timestamp": "2025-01-22T10:30:00Z"
}
}
Status Overview
| Status | When It Happens | What You Do |
|---|
completed | Search finished successfully | Process the product results |
input-required | Need clarification on the brief | Answer the question and continue |
working | Searching across multiple sources | Wait for webhook or poll for updates |
submitted | Custom curation queued | Wait for webhook notification |
failed | Search couldn’t complete | Check error message, adjust brief |
Note: For the complete status list see Task Lifecycle.
Most searches complete immediately. Async processing is only needed for complex scenarios or when the system needs your input.
Next Steps
After discovering products:
- Review Options: Compare products, pricing, and targeting capabilities
- Create Media Buy: Use
create_media_buy to execute campaign
- Prepare Creatives: Use
list_creative_formats to see format requirements
- Upload Assets: Use
sync_creatives to provide creative assets
Learn More