get_plan_audit_logs

Experimental. Campaign governance (sync_plans, check_governance, report_plan_outcome, get_plan_audit_logs) is part of AdCP 3.0 as an experimental surface — it may change between 3.x releases with at least 6 weeks’ notice. Sellers implementing it MUST declare governance.campaign in experimental_features. See experimental status for the full contract.

Retrieve governance state and audit trail for a plan, multiple plans, or an entire portfolio. Returns budget tracking, validation history, and compliance summary.

Request

{
  "tool": "get_plan_audit_logs",
  "arguments": {
    "plan_ids": ["plan_q1_2026_launch"],
    "include_entries": true
  }
}

Results are grouped by governance_context within each plan. Each governed action (media buy, rights license, signal activation, creative service) appears under its governance_context. Multiple plans — retrieve specific plans in one call:

{
  "tool": "get_plan_audit_logs",
  "arguments": {
    "plan_ids": ["plan_q1_2026_launch", "plan_q1_2026_emea"],
    "include_entries": false
  }
}

Portfolio query — retrieve combined audit data for all member plans in one or more portfolios:

{
  "tool": "get_plan_audit_logs",
  "arguments": {
    "portfolio_plan_ids": ["portfolio_nova_brands_2026"],
    "include_entries": true
  }
}

You can combine plan_ids and portfolio_plan_ids to query both specific plans and portfolios in a single call.

Response

{
  "plans": [
    {
      "plan_id": "plan_q1_2026_launch",
      "plan_version": 1,
      "status": "active",
      "budget": {
        "authorized": 500000,
        "committed": 425000,
        "remaining": 75000,
        "utilization_pct": 85
      },
      "channel_allocation": {
        "olv": { "committed": 275000, "pct": 55 },
        "display": { "committed": 150000, "pct": 30 }
      },
      "governed_actions": [
        {
          "governance_context": "gc_mb_seller_456",
          "purchase_type": "media_buy",
          "status": "active",
          "committed": 275000,
          "check_count": 8
        },
        {
          "governance_context": "gc_mb_seller_789",
          "purchase_type": "media_buy",
          "status": "active",
          "committed": 150000,
          "check_count": 7
        },
        {
          "governance_context": "gc_rights_acme_img",
          "purchase_type": "rights_license",
          "status": "active",
          "committed": 75000,
          "check_count": 2
        }
      ],
      "summary": {
        "checks_performed": 14,
        "outcomes_reported": 12,
        "statuses": {
          "approved": 12,
          "denied": 1,
          "conditions": 1,
          "human_reviewed": 1  // also counted in approved or denied above; never counted in conditions (an agent-issued flow-control state, not a human resolution)
        },
        "findings_count": 2,
        "escalations": [
          {
            "check_id": "chk_esc_001",
            "reason": "Budget reallocation exceeds threshold",
            "resolution": "approved_by_human",
            "resolved_at": "2026-03-16T09:30:00Z"
          }
        ],
        "drift_metrics": {
          "escalation_rate": 0.07,
          "escalation_rate_trend": "stable",
          "auto_approval_rate": 0.80,
          "human_override_rate": 0.02,
          "mean_confidence": 0.88,
          "thresholds": {
            "escalation_rate_max": 0.20,
            "escalation_rate_min": 0.02,
            "auto_approval_rate_max": 0.95,
            "human_override_rate_max": 0.15
          }
        }
      },
      "entries": [
        {
          "id": "chk_001",
          "type": "check",
          "timestamp": "2026-03-10T10:05:00Z",
          "caller": "https://orchestrator.pinnacle-media.com/agent",
          "tool": "get_products",
          "check_type": "intent",
          "verdict": "approved",
          "explanation": "Product discovery within budget and channel constraints.",
          "categories_evaluated": ["budget_authority", "strategic_alignment"],
          "policies_evaluated": ["us_coppa"]
        },
        {
          "id": "chk_003",
          "type": "check",
          "timestamp": "2026-03-15T11:05:00Z",
          "caller": "https://ads.seller-example.com/adcp",
          "tool": "create_media_buy",
          "check_type": "execution",
          "mode": "enforce",
          "governance_context": "gc_mb_seller_456",
          "plan_hash": "oR0jFDEtzcwgPbNf-Ofd_fZHYfAyD1TRbzGOFBVCG-c",
          "purchase_type": "media_buy",
          "verdict": "approved",
          "explanation": "Media buy within plan budget ($150,000 of $500,000 remaining). Geo targeting matches authorized markets. COPPA compliance verified.",
          "categories_evaluated": ["budget_authority", "regulatory_compliance", "brand_policy"],
          "policies_evaluated": ["us_coppa", "alcohol_advertising"],
          "findings": [
            {
              "category_id": "budget_authority",
              "severity": "info",
              "explanation": "Budget utilization at 70% after this buy."
            }
          ]
        },
        {
          "id": "out_001",
          "type": "outcome",
          "timestamp": "2026-03-15T11:10:00Z",
          "caller": "https://orchestrator.pinnacle-media.com/agent",
          "governance_context": "gc_mb_seller_456",
          "purchase_type": "media_buy",
          "outcome": "completed",
          "committed_budget": 150000
        },
        {
          "id": "out_del_001",
          "type": "outcome",
          "timestamp": "2026-03-22T00:00:00Z",
          "caller": "https://ads.seller-example.com/adcp",
          "governance_context": "gc_mb_seller_456",
          "purchase_type": "media_buy",
          "outcome": "delivery",
          "outcome_status": "accepted"
        }
      ]
    }
  ]
}

Fields

Request

Field	Type	Required	Description
`plan_ids`	string[]	At least one of `plan_ids`, `portfolio_plan_ids`, or `governance_contexts`	Plan IDs to retrieve.
`portfolio_plan_ids`	string[]	At least one of `plan_ids`, `portfolio_plan_ids`, or `governance_contexts`	Portfolio plan IDs. Expanded to member plans.
`governance_contexts`	string[]	No	Filter results to specific governed actions by their `governance_context` values. When present, only entries and summaries for the matching governed actions are returned.
`purchase_types`	string[]	No	Filter results by purchase type (e.g., `["rights_license"]` to see all rights activity). When present, only entries and summaries matching these purchase types are returned.
`include_entries`	boolean	No	Include the full audit trail. Default: `false`.

Response

Field	Type	Description
`plans`	array	Audit data for each requested plan.
`plans[].plan_id`	string	Plan identifier.
`plans[].plan_version`	number	Current plan version.
`plans[].status`	enum	`active`, `suspended`, or `completed`.
`plans[].budget`	object	Budget state.
`plans[].budget.authorized`	number	Total authorized budget from the plan.
`plans[].budget.committed`	number	Total budget committed from confirmed outcomes.
`plans[].budget.remaining`	number	Authorized minus committed.
`plans[].budget.utilization_pct`	number	Committed as a percentage of authorized.
`plans[].channel_allocation`	object	Current channel mix. Keyed by channel ID.
`plans[].channel_allocation[channel].committed`	number	Budget committed to this channel.
`plans[].channel_allocation[channel].pct`	number	Channel’s share of the authorized total budget.
`plans[].governed_actions`	array	Per-governed-action breakdown, grouped by `governance_context`.
`plans[].governed_actions[].governance_context`	string	Opaque governance context identifying this governed action.
`plans[].governed_actions[].purchase_type`	enum	`media_buy`, `rights_license`, `signal_activation`, or `creative_services`.
`plans[].governed_actions[].status`	enum	`active`, `suspended`, or `completed`.
`plans[].governed_actions[].committed`	number	Budget committed for this governed action.
`plans[].governed_actions[].check_count`	integer	Number of governance checks performed.
`plans[].summary`	object	Aggregate validation and outcome statistics.
`plans[].summary.checks_performed`	number	Total governance checks performed.
`plans[].summary.outcomes_reported`	number	Total outcomes reported.
`plans[].summary.statuses`	object	Count of each governance check status (`approved`, `denied`, `conditions`). Also includes `human_reviewed` as a supplementary count — checks that required human review before resolving. Always a subset of `approved` + `denied`; `conditions` is an agent-issued flow-control state and is never counted here.
`plans[].summary.findings_count`	number	Total findings across all checks and outcomes.
`plans[].summary.escalations`	array	Checks that were escalated for human review and their resolutions.
`plans[].summary.escalations[].check_id`	string	The governance check that was escalated.
`plans[].summary.escalations[].reason`	string	Why the check was escalated.
`plans[].summary.escalations[].resolution`	string	How it was resolved (e.g., `approved_by_human`, `rejected_by_human`).
`plans[].summary.escalations[].resolved_at`	string	ISO 8601 resolution timestamp.
`plans[].summary.drift_metrics`	object	Aggregate governance metrics for detecting oversight drift. See specification.
`plans[].summary.drift_metrics.escalation_rate`	number	Fraction of checks that were escalated for human review (0-1).
`plans[].summary.drift_metrics.escalation_rate_trend`	enum	`increasing`, `stable`, or `declining`.
`plans[].summary.drift_metrics.auto_approval_rate`	number	Fraction of checks approved without human intervention (0-1).
`plans[].summary.drift_metrics.human_override_rate`	number	Fraction of escalations where the human overrode the governance agent’s recommendation (0-1).
`plans[].summary.drift_metrics.mean_confidence`	number	Average confidence score across findings (0-1). Present when findings include confidence.
`plans[].summary.drift_metrics.thresholds`	object	Organization-defined thresholds for drift metrics. When a metric crosses its threshold, the governance agent includes a finding.
`plans[].summary.drift_metrics.thresholds.escalation_rate_max`	number	Maximum acceptable escalation rate. A rate above this may indicate policy miscalibration.
`plans[].summary.drift_metrics.thresholds.escalation_rate_min`	number	Minimum acceptable escalation rate. A rate below this may indicate eroding oversight.
`plans[].summary.drift_metrics.thresholds.auto_approval_rate_max`	number	Maximum acceptable auto-approval rate.
`plans[].summary.drift_metrics.thresholds.human_override_rate_max`	number	Maximum acceptable human override rate.
`plans[].entries`	array	Ordered audit trail (only when `include_entries` is `true`).
`plans[].entries[].id`	string	Entry identifier.
`plans[].entries[].type`	enum	`check` or `outcome`.
`plans[].entries[].timestamp`	string	ISO 8601 timestamp.
`plans[].entries[].plan_id`	string	Plan this entry belongs to. Present when querying multiple plans or a portfolio.
`plans[].entries[].caller`	string	URL of the agent that made the request. Resolved from the credentials used on the governance callback.
`plans[].entries[].tool`	string	The AdCP tool (present for `check` entries).
`plans[].entries[].status`	enum	Governance check status (present for `check` entries).
`plans[].entries[].check_type`	enum	`intent` or `execution` (present for `check` entries). Inferred from the fields present on the original check request.
`plans[].entries[].mode`	enum	`audit`, `advisory`, or `enforce` — governance mode active when this check was evaluated. Recorded by the governance agent from its runtime configuration at check time, not from a plan field. Present on check entries; absent on outcome entries and on governance agents that have not yet adopted this field. Lets auditors distinguish `approved` decisions made under `enforce` from those made under `audit` (where the agent could not have blocked anything).
`plans[].entries[].explanation`	string	Human-readable explanation of the governance decision (present for `check` entries).
`plans[].entries[].policies_evaluated`	array	Registry policy IDs evaluated during this check.
`plans[].entries[].categories_evaluated`	array	Governance categories evaluated (e.g., `budget_authority`, `regulatory_compliance`). Agent-internal labels, not a protocol-level enum — each string is defined by the governance agent’s policy model. Since one governance agent per account composes all specialist review (legal, brand safety, category-specialist reviewers) behind its single endpoint, `categories_evaluated` and `findings[].details` are how internal decomposition surfaces for audit. Consumers MUST treat values as opaque labels: pattern-matching against a fixed list is not safe across agents.
`plans[].entries[].findings`	array	Findings from this check, including category, severity, policy ID, explanation, and confidence.
`plans[].entries[].governance_context`	string	Opaque governance context identifying the governed action this entry belongs to.
`plans[].entries[].plan_hash`	string	Audit-layer binding to the plan revision this attestation was evaluated over — `base64url_no_pad(SHA-256(JCS(plan_payload)))` per Plan binding and audit. Present on `check` entries. Auditors and buyer-side compliance tooling verify by recomputing over the retained plan revision and byte-comparing the decoded 32-byte digests.
`plans[].entries[].purchase_type`	enum	`media_buy`, `rights_license`, `signal_activation`, or `creative_services`. Present when the entry is associated with a specific governed action.
`plans[].entries[].outcome`	enum	Outcome type (present for `outcome` entries).
`plans[].entries[].committed_budget`	number	Budget committed (present for `completed` outcome entries).
`plans[].entries[].outcome_status`	string	Outcome status (present for `outcome` entries).

Authorization

The request schema does not carry an envelope account field — tenant identity is resolved from the submitted IDs. The governance agent MUST verify that the authenticated principal is authorized for every plan_ids member, every plan expanded from portfolio_plan_ids, and every governed action addressed by governance_contexts. When any element fails the authorization check the governance agent MUST fail closed with a generic PLAN_NOT_FOUND response — the error body MUST NOT distinguish “unauthorized” from “not found” or name the offending ID. See Agent and Account Isolation for the pattern and existence-leak guardrails.

Error codes

Code	Recovery	Description
`PLAN_NOT_FOUND`	correctable	No plan with this ID, or the authenticated principal is not authorized for it. The two cases MUST be indistinguishable in the response to prevent plan-ID enumeration.

sync_plans — Push or update plans
check_governance — Validate actions against the plan
report_plan_outcome — Report outcomes to update plan state

Getting Started

Building with AdCP

Protocol

FAQ

Trust & Security

Reference

Certification

get_plan_audit_logs

get_plan_audit_logs

Request

Response

Fields

Request

Response

Authorization

Error codes

​get_plan_audit_logs

​Request

​Response

​Fields

​Request

​Response

​Authorization

​Error codes

​Related tasks

get_plan_audit_logs

Request

Response

Fields

Request

Response

Authorization

Error codes

Related tasks