@raishin/vanguard-frontier-agentic 3.15.0 → 3.17.0

package/tests/agentic-board-suite/fixtures/valid/v11x-designated-challenge-first-answer-over-cap.json

@@ -0,0 +1,598 @@
+{
+  "description": "Valid variant of v11: designated challenge first answer over cap.",
+  "messages": [
+    {
+      "schema_version": "1.0.0",
+      "message_type": "delegate",
+      "task_id": "v11-m1",
+      "correlation_id": "v11-corr",
+      "parent_task_id": null,
+      "hop_count": 0,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:01:00Z",
+      "source": {
+        "kind": "principal",
+        "subject": "subject:corp-idp:7c1e9a40b2d35f68",
+        "role": "platform-owner"
+      },
+      "destination": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "objective": "Assess whether the supplier-onboarding multi-agent design is fit for a non-production pilot.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R2",
+      "data_classification": "internal",
+      "materiality_flags": [],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 0,
+        "limit_tokens": 400000,
+        "spent_tokens": 0,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 0
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "not-required",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "dispatched",
+      "stop_reasons": [],
+      "synthetic": true,
+      "decision_class": "architecture",
+      "decision_owner": {
+        "kind": "principal",
+        "subject": "subject:corp-idp:a94f03d2c8e1b756",
+        "role": "ai-risk-owner"
+      }
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "delegate",
+      "task_id": "v11-m2",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m1",
+      "hop_count": 1,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:02:00Z",
+      "source": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "destination": {
+        "kind": "agent",
+        "id": "agentic-security-review-agent"
+      },
+      "objective": "Challenge the proposal for v11: the design forwards user tokens to the ERP MCP server.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 8,
+        "limit_tokens": 400000,
+        "spent_tokens": 40000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 160
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "not-required",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "dispatched",
+      "stop_reasons": [],
+      "synthetic": true
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "delegate",
+      "task_id": "v11-q1",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m1",
+      "hop_count": 1,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:02:10Z",
+      "source": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "destination": {
+        "kind": "agent",
+        "id": "agentic-security-review-agent"
+      },
+      "objective": "Assess whether the token broker's audience check is reachable from the tool output channel.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 8,
+        "limit_tokens": 400000,
+        "spent_tokens": 40000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 160
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "not-required",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "dispatched",
+      "stop_reasons": [],
+      "synthetic": true
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "delegate",
+      "task_id": "v11-q2",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m1",
+      "hop_count": 1,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:02:20Z",
+      "source": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "destination": {
+        "kind": "agent",
+        "id": "agentic-security-review-agent"
+      },
+      "objective": "Assess whether the session cache can leak credentials across tenants.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 8,
+        "limit_tokens": 400000,
+        "spent_tokens": 40000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 160
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "not-required",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "dispatched",
+      "stop_reasons": [],
+      "synthetic": true
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "challenge",
+      "task_id": "v11-m7",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m2",
+      "hop_count": 2,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:07:00Z",
+      "source": {
+        "kind": "agent",
+        "id": "agentic-security-review-agent"
+      },
+      "destination": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "objective": "Round 2 challenge findings for the revised v11 proposal.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 20,
+        "limit_tokens": 400000,
+        "spent_tokens": 100000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 400
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "not-required",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "stopped",
+      "stop_reasons": [
+        "challenge-rounds-exceeded"
+      ],
+      "synthetic": true,
+      "challenge": {
+        "proposal_task_id": "v11-m2",
+        "round": 3,
+        "hard_gate": true,
+        "outcome": "sustained",
+        "finding": "Third round: the exchange is capped, so the dissent goes to the owner."
+      }
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "escalate",
+      "task_id": "v11-m8",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m7",
+      "hop_count": 3,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:08:00Z",
+      "source": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "destination": {
+        "kind": "principal",
+        "subject": "subject:corp-idp:a94f03d2c8e1b756",
+        "role": "ai-risk-owner"
+      },
+      "objective": "Escalate: the challenge rounds are exhausted.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 20,
+        "limit_tokens": 400000,
+        "spent_tokens": 100000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 400
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "not-required",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "escalated",
+      "stop_reasons": [
+        "challenge-rounds-exceeded"
+      ],
+      "synthetic": true,
+      "escalation": {
+        "reason": "Two rounds did not close the HARD gate.",
+        "route_hint": "Decision owner sends the work back for a decision-request with every round attached."
+      }
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "delegate",
+      "task_id": "v11-m9",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m8",
+      "hop_count": 4,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:09:00Z",
+      "source": {
+        "kind": "principal",
+        "subject": "subject:corp-idp:a94f03d2c8e1b756",
+        "role": "ai-risk-owner"
+      },
+      "destination": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "objective": "Assess and assemble the v11 decision-request with every challenge round and its dissent attached.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 20,
+        "limit_tokens": 400000,
+        "spent_tokens": 100000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 400
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "not-required",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "dispatched",
+      "stop_reasons": [],
+      "synthetic": true
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "decision-request",
+      "task_id": "v11-m10",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m9",
+      "hop_count": 5,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:10:00Z",
+      "source": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "destination": {
+        "kind": "principal",
+        "subject": "subject:corp-idp:a94f03d2c8e1b756",
+        "role": "ai-risk-owner"
+      },
+      "objective": "Request a human decision on the revised v11 proposal with both challenge rounds attached.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 22,
+        "limit_tokens": 400000,
+        "spent_tokens": 110000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 440
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "pending-human-review",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "dispatched",
+      "stop_reasons": [],
+      "synthetic": true,
+      "decision_request": {
+        "decision_class": "architecture",
+        "proposer_agent": "agentic-system-architecture-review-agent",
+        "proposal_summary": "Revised supplier-onboarding design: per-hop audience-bound token exchange replaces passthrough.",
+        "alternatives": [
+          {
+            "kind": "single-agent",
+            "summary": "One agent calling the ERP through its own audience-bound credential.",
+            "rejected_because": "Requires an ERP connector change outside the pilot window."
+          }
+        ],
+        "challenge_task_ids": [
+          "v11-m7"
+        ],
+        "dissent": [
+          {
+            "challenge_task_id": "v11-m7",
+            "position": "Sustained HARD gate carried to the decision owner."
+          }
+        ],
+        "recommended_outcome": "escalate"
+      }
+    },
+    {
+      "schema_version": "1.0.0",
+      "message_type": "decision-record",
+      "task_id": "v11-m11",
+      "correlation_id": "v11-corr",
+      "parent_task_id": "v11-m10",
+      "hop_count": 6,
+      "max_hops": 16,
+      "sent_at": "2026-10-01T09:11:00Z",
+      "source": {
+        "kind": "principal",
+        "subject": "subject:corp-idp:a94f03d2c8e1b756",
+        "role": "ai-risk-owner"
+      },
+      "destination": {
+        "kind": "agent",
+        "id": "agentic-system-architecture-review-agent"
+      },
+      "objective": "Record the decision owner's decision on the revised v11 proposal.",
+      "input_schema_version": "1.0.0",
+      "subject_risk_tier": "R3",
+      "data_classification": "internal",
+      "materiality_flags": [
+        "security-critical"
+      ],
+      "authorized_actions": [
+        "read",
+        "analyze",
+        "recommend",
+        "challenge",
+        "escalate"
+      ],
+      "do_not_do_list": [
+        "Do not execute, deploy, or mutate any system",
+        "Do not request, store, or echo credentials, tokens, or secrets",
+        "Do not record an approval on behalf of a human"
+      ],
+      "evidence_refs": [
+        {
+          "ref": "repo:docs/design/v11.md#section-1",
+          "label": "repo-evidence"
+        }
+      ],
+      "missing_evidence": [],
+      "budget": {
+        "currency": "USD",
+        "limit_tool_calls": 40,
+        "spent_tool_calls": 22,
+        "limit_tokens": 400000,
+        "spent_tokens": 110000,
+        "limit_cost_minor": 2000,
+        "spent_cost_minor": 440
+      },
+      "deadline": "2026-10-08T17:00:00Z",
+      "approval_state": "rejected",
+      "expected_output_schema": "agentic-review-verdict/1.0.0",
+      "status": "completed",
+      "stop_reasons": [],
+      "synthetic": true,
+      "decision_record": {
+        "request_task_id": "v11-m10",
+        "request_digest": "sha256:b72b0e3ee68cdd7aa0b59bb5f42de2ae31e2e8c23eeafdfd228f5f9757c80845",
+        "outcome": "reject",
+        "decided_by": {
+          "kind": "principal",
+          "subject": "subject:corp-idp:a94f03d2c8e1b756",
+          "role": "ai-risk-owner"
+        },
+        "rationale": "Decision taken on the attached proposal, alternatives, challenges and dissent.",
+        "dissent_acknowledged": true,
+        "risk_acceptance_ref": "risk-acceptance:RA-2026-0057",
+        "conditions": [
+          "Keep the per-hop audience-bound token exchange"
+        ]
+      }
+    }
+  ]
+}