Using CPython 3.11.14
Creating virtual environment at: /tmp/dsh-labs-x0suw054/python-env
Activate with: source /tmp/dsh-labs-x0suw054/python-env/bin/activate
Using Python 3.11.14 environment at: /tmp/dsh-labs-x0suw054/python-env
Resolved 5 packages in 68ms
Installed 5 packages in 15ms
 + annotated-types==0.8.0
 + pydantic==2.12.5
 + pydantic-core==2.41.5
 + typing-extensions==4.16.0
 + typing-inspection==0.4.4
npm warn deprecated node-domexception@1.0.0: Use your platform's native DOMException instead

added 545 packages, and audited 546 packages in 31s

65 packages are looking for funding
  run `npm fund` for details

16 vulnerabilities (7 moderate, 9 high)

To address issues that do not require attention, run:
  npm audit fix

To address all issues (including breaking changes), run:
  npm audit fix --force

Run `npm audit` for details.

> baoer-deepseek-harness-labs@0.1.0 check
> tsc --noEmit


> baoer-deepseek-harness-labs@0.1.0 build
> tsc


> baoer-deepseek-harness-labs@0.1.0 test
> node --test --test-concurrency=1 test/*.test.mjs

✔ 六个教学阶段经过真实dsh，搜索只预览，拒绝不泄露全文，预算停请求，关闭后官方reader冷读 (14740.073465ms)
✔ fixture拒绝缺失/错callId的证据，不直接送预写的成功答案 (39.099119ms)
✔ 服务、消费者和prompt贡献随消费者卸载撤销 (44.741349ms)
✔ 服务提供者卸载会停用依赖消费者，再挂载会恢复 (28.442646ms)
✔ JSON provider替换同一service，tools消费者无需改动 (65.230063ms)
✔ 配置/文件数据拒绝重复id、路径id及超长内容 (1.071731ms)
✔ schema拒绝缺失参数，领域验证拒绝路径与未知笔记 (25.496988ms)
✔ pre-execute拒绝时tool body未执行，卸载policy后重新允许 (11.303939ms)
✔ post-execute block不回滚已发生的tool副作用 (9.636317ms)
✔ canonical output校验拒绝不符合声明schema的成功值 (6.510213ms)
✔ 已取消signal在dispatch前阻止body，返回取消结果 (7.483538ms)
✔ waterfall等待下游真实async gate，而非fire-and-forget (20.561486ms)
✔ 读取值是copy，caller无法修改已发布provider快照 (6.52537ms)
✔ 笔记policy拒绝已注册工具时不调用provider，撤销policy后结构化value和展示文本一致 (9.855578ms)
✔ rc.2官方Python源码SDK通过public dsh-bin执行同一实际notes组合 (2092.694374ms)
✔ 真实dsh profile+SDK+Messages adapter完成3次HTTP、2tool并写V4日志 (1186.177277ms)
✔ 模型试图调用未暴露bash不会产生文件副作用 (1067.368011ms)
✔ 共用pre-step预算在2次tool step后拒绝第3次请求 (1243.557289ms)
✔ 同一runtime/session跨run继续；每轮step预算重新计数 (1138.315278ms)
✔ 目标AGENTS/SYSTEM不进入独立sdk-minimal组合的模型请求 (970.368563ms)
✔ provider HTTP401是settled失败，不把空finalResponse与idle当业务成功 (1051.389223ms)
ℹ tests 21
ℹ suites 0
ℹ pass 21
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 25730.901604

> baoer-deepseek-harness-labs@0.1.0 demo
> node dist/offline-demo.js

模型可见历史来自 Session 日志投影；来源：session。
{
  "requests": 3,
  "eventTypes": [
    "agent/inbox/spliced",
    "turn/start",
    "agent/inbox/spliced",
    "step/start",
    "system/message",
    "user/message",
    "request/header",
    "request/context",
    "session/title",
    "session-log-deepseek/delivery-accepted",
    "assistant/message",
    "tool/call",
    "tool/result",
    "step/end",
    "step/start",
    "session-log-deepseek/delivery-accepted",
    "assistant/message",
    "tool/call",
    "tool/result",
    "step/end",
    "step/start",
    "session-log-deepseek/delivery-accepted",
    "assistant/message",
    "step/end",
    "turn/end"
  ],
  "sessionId": "session-35f8d31aefcb44a59b2929d155693a89"
}

> baoer-deepseek-harness-labs@0.1.0 demo:python
> node dist/python-demo.js

{
  "value": {
    "sessionId": "session-97a041da75e64616a2578ab03c9e0536",
    "finalResponse": "Python SDK: source session.",
    "finishReason": "completed",
    "toolResults": 2
  },
  "requests": 3
}
{
  "stage": 0,
  "requests": 1,
  "reason": "completed",
  "taskStatus": "runtime-only",
  "acceptance": {
    "evidenceRead": false,
    "citationValid": false,
    "fixtureAnswerMatches": false
  },
  "finalResponse": "运行时已接收问题。",
  "sessionId": "session-d3101d9c11794de88d502c3c4482222a",
  "artifacts": "/tmp/dsh-labs-x0suw054/examples/artifacts/checkpoint-0"
}
{
  "stage": 1,
  "requests": 3,
  "reason": "completed",
  "taskStatus": "answered-with-evidence",
  "acceptance": {
    "evidenceRead": true,
    "citationValid": true,
    "fixtureAnswerMatches": true
  },
  "finalResponse": "失败 attempt 保留诊断，不进入模型历史。[session]",
  "sessionId": "session-d7d8f454955e44dc939733374429c287",
  "artifacts": "/tmp/dsh-labs-x0suw054/examples/artifacts/checkpoint-1"
}
{
  "stage": 2,
  "requests": 3,
  "reason": "completed",
  "taskStatus": "answered-with-evidence",
  "acceptance": {
    "evidenceRead": true,
    "citationValid": true,
    "fixtureAnswerMatches": true
  },
  "finalResponse": "失败 attempt 保留诊断，不进入模型历史。[session]",
  "sessionId": "session-1700988dd1e840929f848c0f1ba54f4a",
  "artifacts": "/tmp/dsh-labs-x0suw054/examples/artifacts/checkpoint-2"
}
{
  "stage": 3,
  "requests": 3,
  "reason": "completed",
  "taskStatus": "insufficient-evidence",
  "acceptance": {
    "evidenceRead": false,
    "citationValid": false,
    "fixtureAnswerMatches": false
  },
  "finalResponse": "当前政策禁止读取全文，证据不足。",
  "sessionId": "session-ec9c68871ede492ab7beed9849de4bba",
  "artifacts": "/tmp/dsh-labs-x0suw054/examples/artifacts/checkpoint-3"
}
{
  "stage": 4,
  "requests": 2,
  "reason": "blocked",
  "taskStatus": "budget-exhausted",
  "acceptance": {
    "evidenceRead": true,
    "citationValid": false,
    "fixtureAnswerMatches": false
  },
  "finalResponse": "",
  "sessionId": "session-fb396ff262484c418b18cb702a62ecae",
  "artifacts": "/tmp/dsh-labs-x0suw054/examples/artifacts/checkpoint-4"
}
{
  "stage": 5,
  "requests": 3,
  "reason": "completed",
  "taskStatus": "answered-with-evidence",
  "acceptance": {
    "evidenceRead": true,
    "citationValid": true,
    "fixtureAnswerMatches": true
  },
  "finalResponse": "失败 attempt 保留诊断，不进入模型历史。[session]",
  "sessionId": "session-de4d2ed2dd6145c49f229048883744c4",
  "storedEvents": 25,
  "artifacts": "/tmp/dsh-labs-x0suw054/examples/artifacts/checkpoint-5",
  "retainedHome": "/tmp/dsh-labs-x0suw054/examples/artifacts/checkpoint-5/home"
}
