Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
47 changes: 47 additions & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
@@ -1,5 +1,52 @@
# Changelog

## 3.3.0-rc.1

### Minor Changes

- 61a1638: Add the experimental DOOH impression basis (`media_buy.dooh_impression_basis`, RFC #7253). A DOOH `impressions` figure is `plays × audience multiplier`, and the new `enums/impression-estimation-basis.json` says where the multiplier comes from: `currency_measured`, `seller_modeled`, `sensor_measured` or `hybrid`.

- Delivery: `dooh_metrics` carries `impression_basis`, `audience_multiplier` and `basis_metric`, the `vendor_metric_values[]` row a sensor audience comes from. A row mixing tiers reports `impression_basis_breakdown[]` instead: at least two entries with distinct tiers, summing exactly to the row, and no row-level tier or multiplier. `venue_breakdown[]` rows may carry the tier and `plays`, one per screen that played the spot, and no multiplier: a venue's audience per play is its `impressions / plays`.
- Forecast: forecast points, canonical ones included, carry `impression_basis`, `audience_multiplier` and `impression_basis_breakdown[]`, outside `metrics`. A breakdown includes the fallback the seller expects.
- Discovery: `dooh_placement_attributes` carries the planned tier and multiplier.

`impressions` is authoritative and the multiplier is `impressions / plays`, rounded. `hybrid` is a seller model calibrated by sensors on the seller's own screens, never a published currency nor a mix of tiers. All fields are optional. `ooh_metrics.estimation_basis` is unchanged.
- 954eb9d: Campaign governance (experimental `governance.campaign`): settle the rules the specification contradicted or left undefined (adcp#8733, PR 2 of 3). These are normative decisions of the Governance working group. No wire field, enum member or `plan_hash` rule changes. This is a change within the 3.3 pre-release.

- **Budget cut below committed spend.** A `sync_plans` that reduces `budget.total` (or a period `amount`) below the amount already committed is rejected with `VALIDATION_ERROR` and leaves the plan unchanged. Before, the specification said the sync was accepted and flagged on the next check, while the task page said it was rejected.
- **Unapproved seller or undelegated caller on `check_governance`.** A failed `approved_sellers` or `delegations` evaluation for an authenticated caller is returned as a `denied` verdict with a finding and a `check_id`, so the attempt can be audited. This holds in every governance mode. A plan is accessible to its owner principal, to a current delegate, and to a caller that presents a `governance_context` or `consultation_context` issued for that plan. Any other caller receives `PLAN_NOT_FOUND`, the same response as for an unknown `plan_id`, and writes no audit entry. Before, the task page listed `PERMISSION_DENIED` for these cases. `check_governance` still returns `PERMISSION_DENIED` when the caller is unauthenticated or its authenticated identity does not match `caller`. Callers that branch on the error code for the authority cases must also handle the verdict.
- **Human review in every mode.** Audit and advisory mode no longer describe an agent that always returns `approved`: plans with `human_review_required: true` escalate in every mode, and a failed `approved_sellers` or `delegations` evaluation is `denied` in every mode.
- **`plan_id` namespace.** A `plan_id` is unique per owner principal. Two principals may use the same `plan_id`, and a `sync_plans` never reads, changes or reveals another principal's plan.
- **`execute_only`.** A pre-approved action is one whose opaque action binding already holds an approved intent check in the plan's lifecycle. An `execute_only` delegate may act only on such an action, which includes `accept_proposal` for it. The governance agent MUST deny an intent check from an `execute_only` delegate that would create a new binding or increase committed spend.
- **Regions-restricted plans.** An action with no geographic targeting, or with country-level targeting only, is outside a regions-restricted plan: an intent check returns `conditions` naming the region constraint, and an execution check returns `denied`.
- **Unresolvable registry policy ID.** `sync_plans` returns `REFERENCE_NOT_FOUND` when a registry policy ID in `policy_ids` cannot be resolved.
- **Three re-call rule.** "The same action" means re-checks that carry the same `consultation_context`.
- **Explicit `human_review_required: false`.** The `sync_plans` error table states that it is rejected with `VALIDATION_ERROR` on a plan that requires human review.

Training agent: `sync_plans` now rejects a `budget.total` reduced below committed spend, and `check_governance` returns the `denied` verdict for an authenticated execution caller that is not the audience preserved from the approved intent, and for a buyer-side caller whose delegation has ended. It no longer approves a failed `approved_sellers` or `delegations` evaluation in advisory or audit mode. For a plan that does not exist or that the caller cannot access, `check_governance` now returns the `PLAN_NOT_FOUND` error; it returned a `denied` verdict with a `plan_lookup` finding.
- 3a3bf1e: Withdraw `metric_id` from performance standards and restore the `evaluator` value of `x-entity-types` as deprecated (adcp#8501, entries 6 and 26). A minor release doesn't change what a signed term means or remove an existing enum value (DR-0036).

**`metric_id` on performance standards (#8436, `3.3.0-beta.1` through `3.3.0-rc.0`).** Removed from `core/canonical-performance-standard.json`, `core/performance-standard.json` and the `required_performance_standards` entries of `core/product-offer-filters.json`, with the guard that kept it off `viewability` and `completion_rate`. The three schemas and the `performance-standard-metric` description return to their 3.2.8 form. The field selects which vendor value decides a signed performance standard, so a 3.2 buyer would read the same standard differently. It is planned for 4.0 as `performance_standard_metric_id_required_in_4_0` in the 4.0 compatibility registry, which keeps the design.

**`evaluator` in `x-entity-types`.** Restored with its definition and listed in `x-deprecated-enum-values`. The 3.2.8 versioning page says "Existing values are never removed or renamed." The removal of the experimental `evaluator_id` source form (#8045) stands, so no 3.3 field carries the value.

These are beta-to-beta changes, not changes to a stable release: 3.2.8 has no `metric_id` on performance standards. The betas are immutable, so a seller or SDK built from a beta bundle stops sending and emitting the field. `scripts/lint-additive-minor-baseline.json` drops the two rows these entries covered and is now empty.

### Patch Changes

- d24319d: Campaign governance (experimental `governance.campaign`): follow-up errata found during the coherence sweep (adcp#8733, PR 4). Each edit follows a rule that is already stated elsewhere in the specification or schemas. No schema property, enum member or `plan_hash` rule changes.

- **Human-review trigger list.** The trust page, known limitations and the FAQ name all five schema triggers: `fair_housing`, `fair_lending`, `fair_employment` and `pharmaceutical_advertising` in `policy_categories`, and `eu_ai_act_annex_iii` in `policy_ids`. The FAQ no longer describes 3.0 in the future tense, and says that an omitted flag is accepted from 3.3.
- **Who sets the flag.** The `requires_human_review` descriptions in `policy-category-definition.json` and `policy-entry.json` say the governance agent sets `plan.human_review_required`, as the specification does. The category description no longer says that a policy can override the category's setting: a policy adds the requirement and cannot remove it.
- **Reallocation fields.** `budget.reallocation_threshold` is no longer documented as always required. The schema requires exactly one of `reallocation_threshold` or `reallocation_unlimited: true`, and the `sync_plans` field table now lists both. The reallocation table and the `reallocation_threshold` description point to `reallocation_unlimited: true` for unlimited reallocation, because a threshold equal to `total` tightens when `total` changes.
- **Overview page.** Advisory mode is described as returning `approved` with findings, and the "Execute" step and the image description no longer say budget is reserved.
- **Audit logs.** The audit-trail page and the `entries[].mode` description no longer refer to a closed issue or to mode as a plan property. The `entries[].mode` description no longer mentions a `governed_actions[].mode` field that does not exist. The specification and the audit-trail page now state the same strength as the schema: a governance agent SHOULD record the mode on check entries, and outcome entries carry none.
- **Mode on the response.** The specification no longer says that governance mode is "not a protocol-level field". The `check_governance` response has carried an optional `mode` property since 3.1, and the task page documents it. The specification now says that mode is agent configuration, that no request carries it, and that the agent may report it on the response. Callers still MUST NOT branch on it.
- **Delegation expiry.** The `delegations[].expires_at` description said that the governance agent denies an expired delegate's actions. It now follows the plan-access rule: a re-check that presents a context issued for the plan, and an execution check on an action the agent proposed, are `denied`, and any other check from that agent returns `PLAN_NOT_FOUND`.

Training agent: the `mode` property on its `sync_plans` tool is described as a sandbox test control. Its behaviour is unchanged.
- c925886: Docs: `@adcp/sdk` 17.0.0-beta.29 and Python `adcp` 10.0.0b20 bundle `3.3.0-rc.0`, and the public test agent's opt-in 3.3 preview moves to `3.3-rc.0`. Choose your SDK, the 3.3 prerelease page, the release notes, Versions, and the quickstart name the new versions and pin. `@adcp/sdk` 17.0.0-beta.24 through 17.0.0-beta.28 and Python `adcp` 10.0.0b17 through 10.0.0b19 support beta.6. The training agent's hosted preview is `3.3-rc.0`, validated against the published `3.3.0-rc.0` schemas.

## 3.3.0-rc.0

### Minor Changes
Expand Down
196 changes: 196 additions & 0 deletions dist/compliance/3.3.0-rc.1/domains/brand/index.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,196 @@
id: brand_baseline
version: "1.0.0"
title: "Brand baseline"
protocol: brand
category: brand_baseline
summary: "Baseline protocol storyboard — every brand agent must declare the brand protocol in capabilities and return a schema-valid brand identity."
track: brand
required_tools:
- get_brand_identity

narrative: |
Brand protocol agents are the identity layer of AdCP. Their job is to hold
brand identity data (names, logos, colors, fonts, tone) and expose it to
other agents — buyer agents, creative agents, DSPs — that need to render
on-brand creative or verify who a campaign is for.

The baseline tests the minimum contract that every brand agent honors,
regardless of what additional capabilities (rights licensing, creative
approval) it layers on top:

1. Declare `brand` in `supported_protocols` on `get_adcp_capabilities`.
2. Respond to `get_brand_identity` with a schema-valid identity manifest.
3. Reject unknown `brand_id` values with a structured error.

Rights licensing (`get_rights`, `acquire_rights`, `update_rights`,
`creative_approval`) ships experimentally in 3.0 and is covered by the
`brand-rights` specialism storyboard, not this baseline.

agent:
interaction_model: brand_agent
capabilities: []
examples:
- "Any brand agent (simple identity host or full rights platform)"
- "Brand-owned agents (Acme Outdoor)"
- "Third-party brand identity platforms"
- "Agency-hosted brand agents"

caller:
role: buyer_agent
example: "Any buyer, creative agent, or DSP needing brand identity"

prerequisites:
description: |
The test kit provides a sample brand (Nova Motors) that any brand agent
can serve identity for.
test_kit: "test-kits/nova-motors.yaml"

phases:
- id: capability_discovery
title: "Capability discovery"
narrative: |
The buyer calls `get_adcp_capabilities` to confirm the agent declares
the brand protocol before issuing any brand-identity call.

steps:
- id: get_capabilities
title: "Check agent capabilities"
narrative: |
Verify that the agent declares `brand` in `supported_protocols`.
Without this claim the buyer MUST NOT send `get_brand_identity`.
task: get_adcp_capabilities
schema_ref: "protocol/get-adcp-capabilities-request.json"
response_schema_ref: "protocol/get-adcp-capabilities-response.json"
doc_ref: "/protocol/get_adcp_capabilities"
comply_scenario: capability_discovery
stateful: false
expected: |
Return capabilities declaring `brand` in `supported_protocols`.

sample_request:
context:
correlation_id: "brand_baseline--get_capabilities"
validations:
- check: response_schema
description: "Response matches get-adcp-capabilities-response.json schema"
- check: field_present
path: "supported_protocols"
description: "Response declares supported_protocols"

- id: brand_identity_retrieval
title: "Brand identity retrieval"
narrative: |
The buyer calls `get_brand_identity` to retrieve the brand's identity
manifest. The minimum contract is a schema-valid response that echoes
the requested `brand_id` and carries at least one name.

steps:
- id: get_brand_identity
title: "Retrieve brand identity"
narrative: |
The buyer calls `get_brand_identity` with a known `brand_id`. The
response MUST match the brand-identity schema and echo the
requested `brand_id`. Rich fields (logos, colors, fonts, tone,
visual_guidelines) are optional at the baseline level — the
minimum bar is that identity resolution works and is schema-valid.
task: get_brand_identity
schema_ref: "brand/get-brand-identity-request.json"
response_schema_ref: "brand/get-brand-identity-response.json"
doc_ref: "/brand-protocol/tasks/get_brand_identity"
stateful: false
expected: |
Return a schema-valid brand identity that echoes the requested
brand_id and includes at least one name.

sample_request:
brand_id: "nova_motors"
context:
correlation_id: "brand_baseline--get_brand_identity"
context_outputs:
- path: "brand_id"
key: "brand_id"

validations:
- check: response_schema
description: "Response matches get-brand-identity-response.json schema"
- check: field_present
path: "brand_id"
description: "Response includes brand_id"
- check: field_value
path: "brand_id"
value: "nova_motors"
description: "Returned brand_id echoes the requested brand"
- check: field_present
path: "names"
description: "Response includes brand names"

- id: unknown_brand_rejection
title: "Unknown brand rejection"
narrative: |
Agents MUST reject unknown `brand_id` values with a structured
AdCP error rather than returning an empty or fabricated manifest.

steps:
- id: get_brand_identity_unknown
title: "Reject unknown brand ID"
narrative: |
The buyer calls `get_brand_identity` with a `brand_id` the agent
does not serve. The response MUST be a structured error with a
recovery classification — not a success response with empty
fields.
task: get_brand_identity
schema_ref: "brand/get-brand-identity-request.json"
response_schema_ref: "brand/get-brand-identity-response.json"
doc_ref: "/brand-protocol/tasks/get_brand_identity"
stateful: false
expected: |
Return an AdCP error response indicating the brand is not known
to this agent.

sample_request:
brand_id: "brand_that_does_not_exist_12345"
context:
correlation_id: "brand_baseline--get_brand_identity_unknown"

expect_error: true
negative_path: payload_well_formed
validations:
- check: error_code
value: "REFERENCE_NOT_FOUND"
description: "Error code indicates brand-not-found. REFERENCE_NOT_FOUND is the canonical fallback per error-handling.mdx (brands lack a dedicated *_NOT_FOUND code)."

- id: brand_discovery
title: "Optional brand discovery"
narrative: |
Agents that implement the experimental `search_brands` task expose a
schema-valid discovery surface without making that task part of the
baseline brand protocol requirement.

steps:
- id: search_brands
title: "Search the agent's brand roster"
task: search_brands
schema_ref: "brand/search-brands-request.json"
response_schema_ref: "brand/search-brands-response.json"
doc_ref: "/brand-protocol/tasks/search_brands"
comply_scenario: brand_discovery
requires_tool: search_brands
stateful: false
expected: |
Return a schema-valid ranked array of brand stubs. An empty array is
valid when the agent has no matching roster entry.
sample_request:
query: "automotive brands with electric vehicles"
countries: ["US"]
context:
correlation_id: "brand_baseline--search_brands"
validations:
- check: response_schema
description: "Response matches search-brands-response.json schema"
- check: field_present
path: "brands"
description: "Discovery response includes the brands array"
- check: field_value
path: "context.correlation_id"
value: "brand_baseline--search_brands"
description: "Context correlation_id returned unchanged"
Loading
Loading