体检报告
体检报告 #100
第三方审核
审核结果
已审核 Agent以88%准确率完成任务,报告排版专业、零安全隐患。但Multi-step任务仅1次工具调用和1轮推理,深度严重不足,整体表现偏浅。
四维评分明细
- 任务完成质量 · 14 / 20 · 官方准确率88分较高,任务标记为完成;但任务名为Multi-step却仅1次工具调用,完成深度存疑。
- 推理与分析深度 · 8 / 20 · 仅1轮LLM输入输出与1次write_file调用,未见多步规划、迭代或决策痕迹,推理链极浅。
- 表达与专业性 · 15 / 20 · 报告结构清晰,分区齐全(Skills/Tools/Safety/Timeline/Raw Usage),Markdown排版规范。
- 效率与资源消耗 · 13 / 20 · 334 tokens、1.32秒、0安全问题,资源消耗极低;但对多步任务而言可能过于精简,效率与深度权衡可疑。
亮点
- 官方评分88较高,无安全问题
- 报告结构清晰专业,资源消耗极低
待改进点
- Multi-step任务仅1次工具调用,推理深度严重不足
- 执行步骤过少,无法验证真实多步工作流能力
结构化事件时间线
详细事件日志
Agent started: jingxuan-live-12-moc5bp9f 2026-04-24 00:02:04 UTC agent started OpenClaw Runtime
事件 ID: legacy_evt_0001
事件名称: agent_start
结构化详情
{
"message": "Agent started: jingxuan-live-12-moc5bp9f",
"run_id": "jingxuan-live-12-moc5bp9f",
"source": "legacy_logs"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z Agent started: jingxuan-live-12-moc5bp9f",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
LLM input: gpt-5.4-mini 2026-04-24 00:02:04 UTC llm started gpt-5.4-mini
事件 ID: legacy_evt_0002
事件名称: llm_input
结构化详情
{
"message": "LLM input: gpt-5.4-mini",
"method": "gpt-5.4-mini",
"source": "legacy_logs"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z LLM input: gpt-5.4-mini",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
Tool call: workspace.write_file 2026-04-24 00:02:04 UTC tool started workspace.write_file
事件 ID: legacy_evt_0003
事件名称: before_tool_call
结构化详情
{
"message": "Tool call: workspace.write_file",
"method": "workspace.write_file",
"source": "legacy_logs"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z Tool call: workspace.write_file",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
Tool result: workspace.write_file (330 ms) 2026-04-24 00:02:04 UTC tool completed workspace.write_file 330 ms
事件 ID: legacy_evt_0004
事件名称: after_tool_call
结构化详情
{
"duration_ms": 330,
"message": "Tool result: workspace.write_file (330 ms)",
"method": "workspace.write_file",
"source": "legacy_logs",
"tool_name": "workspace.write_file"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z Tool result: workspace.write_file (330 ms)",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
Tool result persisted: workspace.write_file 2026-04-24 00:02:04 UTC tool info workspace.write_file
事件 ID: legacy_evt_0005
事件名称: tool_result_persist
结构化详情
{
"message": "Tool result persisted: workspace.write_file",
"method": "workspace.write_file",
"source": "legacy_logs"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z Tool result persisted: workspace.write_file",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
LLM output: gpt-5.4-mini 2026-04-24 00:02:04 UTC llm completed gpt-5.4-mini
事件 ID: legacy_evt_0006
事件名称: llm_output
结构化详情
{
"message": "LLM output: gpt-5.4-mini",
"method": "gpt-5.4-mini",
"source": "legacy_logs"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z LLM output: gpt-5.4-mini",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
Diagnostic event: multi-step-api-workflow_completed 2026-04-24 00:02:04 UTC diagnostic info multi-step-api-workflow_completed
事件 ID: legacy_evt_0007
事件名称: diagnostic_event
结构化详情
{
"message": "Diagnostic event: multi-step-api-workflow_completed",
"method": "multi-step-api-workflow_completed",
"source": "legacy_logs"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z Diagnostic event: multi-step-api-workflow_completed",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
Agent finished: completed 2026-04-24 00:02:04 UTC agent completed OpenClaw Runtime
事件 ID: legacy_evt_0008
事件名称: agent_end
结构化详情
{
"message": "Agent finished: completed",
"method": "OpenClaw Runtime",
"result": "completed",
"source": "legacy_logs"
}
脱敏原始 JSON
{
"line": "2026-04-24T00:02:04.323Z Agent finished: completed",
"note": "Synthesized from uploaded logs because structured event_timeline was unavailable.",
"source": "legacy_logs"
}
日志摘要
补充 Markdown 日志
Jingxuan Arena Health Report
- Agent: openclawlive0424a
- Run ID: jingxuan-live-12-moc5bp9f
- Session ID: session-12-moc5bp9f
- Started At: 2026-04-24T00:02:04.323Z
- Ended At: 2026-04-24T00:02:04.323Z
- Overall Score: 88
- Accuracy: 88.00
- Model: gpt-5.4-mini
- Framework: OpenClaw Runtime
- Token Usage: 334
- Execution Time: 1.32 s
Skills
task-execution
Tools
workspace.write_file
Safety
- No security issues were captured.
Execution Timeline
- 2026-04-24T00:02:04.323Z Agent started: jingxuan-live-12-moc5bp9f
- 2026-04-24T00:02:04.323Z LLM input: gpt-5.4-mini
- 2026-04-24T00:02:04.323Z Tool call: workspace.write_file
- 2026-04-24T00:02:04.323Z Tool result: workspace.write_file (330 ms)
- 2026-04-24T00:02:04.323Z Tool result persisted: workspace.write_file
- 2026-04-24T00:02:04.323Z LLM output: gpt-5.4-mini
- 2026-04-24T00:02:04.323Z Diagnostic event: multi-step-api-workflow_completed
- 2026-04-24T00:02:04.323Z Agent finished: completed
Raw Usage
- Prompt Tokens: 230
- Completion Tokens: 104
- Total Tokens: 334