活动合约

查询期望的字段

字段	类型	为什么存在
timestamp_utc	Timestamp	终端智能体运行时间（UTC）。
workflow	Utf8	稳定的产品工作流程或任务类别。
model	Utf8	从提供商响应中记录的模型标识符。
status	Utf8	技术运行成功或失败。
reviewer_outcome	Utf8	接受、修改、拒绝或 not_reviewed。
human_handoff	Boolean	是否由人类接管该任务。
duration_ms	Float64	端到端工作流程持续时间。
estimated_cost_usd	Float64	运行的标准化估计模型成本。

DataFusion SQL

复制查询

sql

SELECT
  workflow,
  model,
  COUNT(*) AS runs,
  100.0 * SUM(CASE WHEN status = 'success' THEN 1 ELSE 0 END)
    / NULLIF(COUNT(*), 0) AS technical_success_rate_pct,
  100.0 * SUM(CASE WHEN reviewer_outcome = 'accepted' THEN 1 ELSE 0 END)
    / NULLIF(SUM(CASE
      WHEN reviewer_outcome <> 'not_reviewed' THEN 1 ELSE 0
    END), 0) AS reviewed_acceptance_rate_pct,
  100.0 * SUM(CASE WHEN human_handoff THEN 1 ELSE 0 END)
    / NULLIF(COUNT(*), 0) AS human_handoff_rate_pct,
  AVG(duration_ms) AS average_duration_ms,
  SUM(estimated_cost_usd) / NULLIF(COUNT(*), 0) AS cost_per_run_usd
FROM agent_run_outcomes
WHERE timestamp_utc >= now() - INTERVAL '30 days'
GROUP BY workflow, model
HAVING COUNT(*) >= 20
ORDER BY reviewed_acceptance_rate_pct, runs DESC;

此只读查询是针对空类型表计划和执行的阿帕奇 DataFusion 45.2.0。确定性样本输出是综合的并单独审查；根据您自己的数据验证字段类型、阈值和业务定义。阅读测试方法。

查询结果

按工作流程审查智能体接受情况

退款审核在技术上完成，但比其他工作流程需要更多的修改或人工接管。

比较数字结果字段

support_triage

86.4%

comparison 18.2%

refund_review

68.7%

comparison 42.9%

knowledge_draft

91.6%

comparison 8.4%

workflow	model	runs	technical_success_rate_pct	reviewed_acceptance_rate_pct	human_handoff_rate_pct	average_duration_ms	cost_per_run_usd
support_triage	configured-model-a	1,840	98.1	86.4	18.2	4,820	0.04
refund_review	configured-model-b	620	96.8	68.7	42.9	8,110	0.07
knowledge_draft	configured-model-a	940	99.2	91.6	8.4	3,910	0.04