Orient, select an agent, and delegate bounded work

workflow-orient-delegate

The agent gathers task and actor context before creating exactly one scoped child.

Prompt

First read your task context and list company agents. Then create exactly one high-priority child task titled Audit protocol coverage assigned to actor-manager. Do not finish the parent task.

Fixture

../fixtures/companies/orientation.json

Authority

{
  "actorId": "actor-engineer",
  "capabilities": [
    "discovery:agents:read",
    "delegation:tasks:create"
  ],
  "explicitClaims": [
    "discovery:agents:read",
    "delegation:tasks:create"
  ],
  "taskId": "task-orientation"
}

Current results

Model	Configuration	Result	Estimated cost	Run time
claude-sonnet-5	live-acpx-0.13.1-claude-acp-0.70.0-sonnet-5	INFRA	unknown	2026-09-06T20:52:28.782Z
claude-sonnet-5	live-claude-managed-agents-2026-04-01-sonnet-5	PASS	$0.022036 estimated	2026-09-06T20:54:28.327Z
global.anthropic.claude-sonnet-4-6	live-aws-agentcore-global-anthropic-claude-sonnet-4-6	INFRA	unknown	2026-09-06T20:53:16.800Z
gpt-5.4-mini	live-codex-pinned	INFRA	unknown	unknown
gpt-5.6-sol	live-acpx-0.13.1-codex-acp-1.6.2-control	Not run	—	—
openrouter/anthropic/claude-sonnet-5	live-opencode-1.18.17-openrouter-claude-sonnet-5	PASS	$0.014655 estimated	2026-09-06T20:55:25.778Z
openrouter/deepseek/deepseek-v4-flash-0731	live-opencode-1.18.17-openrouter-deepseek-v4-flash-0731	PASS	$0.000360 estimated	2026-09-06T20:58:24.827Z
openrouter/google/gemini-3.8-flash	live-opencode-1.18.17-openrouter-gemini-3.8-flash	PASS	$0.011735 estimated	2026-09-06T20:55:35.372Z
openrouter/openai/gpt-6-astra	live-opencode-1.18.17-openrouter-gpt-6-astra	PASS	$0.036614 estimated	2026-09-06T20:57:04.698Z
openrouter/qwen/qwen3.8-max-0902	live-opencode-1.18.17-openrouter-qwen3.8-max-0902	PASS	$0.019526 estimated	2026-09-06T20:57:56.022Z
openrouter/z-ai/glm-5.3	live-opencode-1.18.17-openrouter-glm-5.3	FAIL	$0.165860 estimated	2026-09-06T20:56:04.527Z

Assertions

context

semantic_operation

{
  "id": "context",
  "kind": "semantic_operation",
  "minCalls": 1,
  "operationId": "get_task_context",
  "requireSuccess": true
}

agents

semantic_operation

{
  "id": "agents",
  "kind": "semantic_operation",
  "minCalls": 1,
  "operationId": "list_agents",
  "requireSuccess": true
}

child

semantic_operation

{
  "id": "child",
  "kind": "semantic_operation",
  "maxCalls": 1,
  "minCalls": 1,
  "operationId": "create_task",
  "requireSuccess": true
}

context-before-child

operation_order

{
  "after": "create_task",
  "before": "get_task_context",
  "id": "context-before-child",
  "kind": "operation_order"
}

agents-before-child

operation_order

{
  "after": "create_task",
  "before": "list_agents",
  "id": "agents-before-child",
  "kind": "operation_order"
}

one-task

state_effect

{
  "collection": "tasks",
  "id": "one-task",
  "kind": "state_effect",
  "maximumAdded": 1,
  "minimumAdded": 1
}

no-finish

semantic_operation

{
  "id": "no-finish",
  "kind": "semantic_operation",
  "maxCalls": 0,
  "operationId": "finish_task"
}

terminal-completed

terminal

{
  "id": "terminal-completed",
  "kind": "terminal",
  "status": "completed"
}

mock-only

network

{
  "id": "mock-only",
  "kind": "network",
  "realPaperclipRequests": 0
}

Full authored test definition

{
  "authority": {
    "actorId": "actor-engineer",
    "capabilities": [
      "discovery:agents:read",
      "delegation:tasks:create"
    ],
    "explicitClaims": [
      "discovery:agents:read",
      "delegation:tasks:create"
    ],
    "taskId": "task-orientation"
  },
  "checks": [
    {
      "id": "context",
      "kind": "semantic_operation",
      "minCalls": 1,
      "operationId": "get_task_context",
      "requireSuccess": true
    },
    {
      "id": "agents",
      "kind": "semantic_operation",
      "minCalls": 1,
      "operationId": "list_agents",
      "requireSuccess": true
    },
    {
      "id": "child",
      "kind": "semantic_operation",
      "maxCalls": 1,
      "minCalls": 1,
      "operationId": "create_task",
      "requireSuccess": true
    },
    {
      "after": "create_task",
      "before": "get_task_context",
      "id": "context-before-child",
      "kind": "operation_order"
    },
    {
      "after": "create_task",
      "before": "list_agents",
      "id": "agents-before-child",
      "kind": "operation_order"
    },
    {
      "collection": "tasks",
      "id": "one-task",
      "kind": "state_effect",
      "maximumAdded": 1,
      "minimumAdded": 1
    },
    {
      "id": "no-finish",
      "kind": "semantic_operation",
      "maxCalls": 0,
      "operationId": "finish_task"
    },
    {
      "id": "terminal-completed",
      "kind": "terminal",
      "status": "completed"
    },
    {
      "id": "mock-only",
      "kind": "network",
      "realPaperclipRequests": 0
    }
  ],
  "description": "The agent gathers task and actor context before creating exactly one scoped child.",
  "fixture": "../fixtures/companies/orientation.json",
  "id": "workflow-orient-delegate",
  "prompt": "First read your task context and list company agents. Then create exactly one high-priority child task titled Audit protocol coverage assigned to actor-manager. Do not finish the parent task.",
  "schema": "paperclip-runner/eval-case/v1",
  "title": "Orient, select an agent, and delegate bounded work"
}