agent-observability

DeepSeek Harness 的执行迷宫——看 Agent 真实怎么干活：迷宫时间轴 · 数据轨道 · 确定性执行分析 · 多会话对比 | The execution maze for DSH agents: maze timeline, per-step data tracks, deterministic execution analysis, multi-session comparison. Formerly dsh-trace-compare.

Read-only Agent work-path observer for DeepSeek Harness

OpenTelemetry tracing for DeepSeek Harness (dsh): turns each agent turn into a GenAI span tree — steps, LLM calls with TTFT, tool executions, token usage — exported over standard OTLP to Jaeger, Grafana Tempo, SigNoz, Langfuse, or any compatible backend.

tencentcloud-agentobs-sdk-dsh 是一个 DeepSeek Harness (DSH) 可观测插件，直接将 GenAI trace 数据上报到腾讯云日志服务 (CLS)。 它观察 DSH 原生的 session、agent loop、LLM stream 和 tool 生命周期，将其转换为 腾讯云AI Agent可观测规范的 5 层 span 层级模型（entry → agent → step → chat → tool），并通过 tencentcloud-cls-sdk-js 直接上报到 CLS，无需额外部署 OTLP 收集器或 sidecar。

Explain failed DeepSeek Harness agent runs locally, then produce a redacted recovery plan. No model required.

Session execution analytics and audit reports for DeepSeek Harness — see how your agent actually worked

Actionable quality diagnostics for DSH agent sessions, with tool-call evidence and next-step recommendations.

Know when your DeepSeek Harness tasks need you. Local alerts, quiet notifications, and a task inbox.

Mission Control dashboard and human-review workflow for DeepSeek Harness sessions, turns, and tool activity.

DeepSeek Harness 会话回放与脱敏分享插件：将真实 Agent 轨迹导出为独立交互 HTML，用于文档、演示和问题反馈。

Task flow / execution graph plugin for DeepSeek Harness (DSH) — visualize a single task's full run: agents, tools, skills, subtasks, retries, live status. DSH 单任务执行流程图谱插件。

DeepSeek Harness（DSH）只读执行轨迹复盘插件，支持规则分析、独立模型解读、证据定位、任务概览与运行对比。

Controlled request-surface replay and regression workbench for DeepSeek Harness

DSH 插件：Jupyter 式活体运行本——回合可回放的数据流 DAG，含 git 提交出处节点、子 agent 交接边、持久账本、悬停重跑与 LLM 解释

Local JSONL run tracing plugin for DeepSeek Harness — records agent runs, model steps, tool calls, timings, outcomes, and token usage.

DSH-Plugin: structural observability layer - real tool/skill usage recording (tools/result), skill-catalog audit, watchdog. No model self-reporting.

Read-only live subagent process inspector for DeepSeek Harness Web

DSH plugin for agent observability and security evaluation

Design, observe, diagnose, and reuse evidence-backed multi-Agent workflows inside DeepSeek Harness

DeepSeek Harness Litefuse Plugin for Agent Observability and Evals

DSH-Plugin: per-reply skill trail - truthful per-turn triggered/used tool & skill badges, folded from the session log (no model self-reporting)