paperclip-runner evalsstatic evidence

← Overview

Orient, request approval, and schedule continuation

workflow-governed-wait

Governed work creates one approval and one bounded wake without premature completion.

Prompt

Read your task context. Request one production_release approval with payload environment set to production, then schedule one approval_resolved wake 5 ticks from now with payload approval set to pending. Do not finish the task.

Fixture

../fixtures/companies/orientation.json

Authority

{
  "actorId": "actor-engineer",
  "capabilities": [
    "governance:approvals:request",
    "control_plane:wakes"
  ],
  "explicitClaims": [
    "governance:approvals:request",
    "control_plane:wakes"
  ],
  "taskId": "task-orientation"
}

Current results

ModelConfigurationResultEstimated costRun time
claude-sonnet-5live-acpx-0.13.1-claude-acp-0.70.0-sonnet-5INFRAunknown2026-09-06T20:52:35.281Z
claude-sonnet-5live-claude-managed-agents-2026-04-01-sonnet-5PASS$0.011728 estimated2026-09-06T20:54:28.576Z
global.anthropic.claude-sonnet-4-6live-aws-agentcore-global-anthropic-claude-sonnet-4-6PASS$0.046431 estimated2026-09-06T20:53:21.925Z
gpt-5.4-minilive-codex-pinnedINFRAunknownunknown
gpt-5.6-sollive-acpx-0.13.1-codex-acp-1.6.2-controlINFRAunknown2026-09-06T20:52:24.813Z
openrouter/anthropic/claude-sonnet-5live-opencode-1.18.17-openrouter-claude-sonnet-5PASS$0.011432 estimated2026-09-06T20:55:26.107Z
openrouter/deepseek/deepseek-v4-flash-0731live-opencode-1.18.17-openrouter-deepseek-v4-flash-0731PASS$0.000715 estimated2026-09-06T20:58:24.629Z
openrouter/google/gemini-3.8-flashlive-opencode-1.18.17-openrouter-gemini-3.8-flashPASS$0.024668 estimated2026-09-06T20:55:35.590Z
openrouter/openai/gpt-6-astralive-opencode-1.18.17-openrouter-gpt-6-astraPASS$0.045872 estimated2026-09-06T20:57:04.166Z
openrouter/qwen/qwen3.8-max-0902live-opencode-1.18.17-openrouter-qwen3.8-max-0902PASS$0.008512 estimated2026-09-06T20:57:55.866Z
openrouter/z-ai/glm-5.3live-opencode-1.18.17-openrouter-glm-5.3PASS$0.093828 estimated2026-09-06T20:56:12.863Z

Assertions

context
semantic_operation
{
  "id": "context",
  "kind": "semantic_operation",
  "minCalls": 1,
  "operationId": "get_task_context",
  "requireSuccess": true
}
approval
semantic_operation
{
  "id": "approval",
  "kind": "semantic_operation",
  "maxCalls": 1,
  "minCalls": 1,
  "operationId": "request_approval",
  "requireSuccess": true
}
wake
semantic_operation
{
  "id": "wake",
  "kind": "semantic_operation",
  "maxCalls": 1,
  "minCalls": 1,
  "operationId": "schedule_wake",
  "requireSuccess": true
}
context-before-approval
operation_order
{
  "after": "request_approval",
  "before": "get_task_context",
  "id": "context-before-approval",
  "kind": "operation_order"
}
approval-before-wake
operation_order
{
  "after": "schedule_wake",
  "before": "request_approval",
  "id": "approval-before-wake",
  "kind": "operation_order"
}
one-approval
state_effect
{
  "collection": "approvals",
  "id": "one-approval",
  "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": [
      "governance:approvals:request",
      "control_plane:wakes"
    ],
    "explicitClaims": [
      "governance:approvals:request",
      "control_plane:wakes"
    ],
    "taskId": "task-orientation"
  },
  "checks": [
    {
      "id": "context",
      "kind": "semantic_operation",
      "minCalls": 1,
      "operationId": "get_task_context",
      "requireSuccess": true
    },
    {
      "id": "approval",
      "kind": "semantic_operation",
      "maxCalls": 1,
      "minCalls": 1,
      "operationId": "request_approval",
      "requireSuccess": true
    },
    {
      "id": "wake",
      "kind": "semantic_operation",
      "maxCalls": 1,
      "minCalls": 1,
      "operationId": "schedule_wake",
      "requireSuccess": true
    },
    {
      "after": "request_approval",
      "before": "get_task_context",
      "id": "context-before-approval",
      "kind": "operation_order"
    },
    {
      "after": "schedule_wake",
      "before": "request_approval",
      "id": "approval-before-wake",
      "kind": "operation_order"
    },
    {
      "collection": "approvals",
      "id": "one-approval",
      "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": "Governed work creates one approval and one bounded wake without premature completion.",
  "fixture": "../fixtures/companies/orientation.json",
  "id": "workflow-governed-wait",
  "prompt": "Read your task context. Request one production_release approval with payload environment set to production, then schedule one approval_resolved wake 5 ticks from now with payload approval set to pending. Do not finish the task.",
  "schema": "paperclip-runner/eval-case/v1",
  "title": "Orient, request approval, and schedule continuation"
}