agent-observability 8
The execution maze for DeepSeek Harness agents: see how the agent actually worked — main path, detours and backtracks on one timeline, with data tracks, deterministic analysis, and multi-session comparison
dsh plugin add dsh-mazeKnow when your DeepSeek Harness tasks need you. Local alerts, quiet notifications, and a task inbox.
dsh plugin add dsh-deepcanaryWhat a DeepSeek Harness run costs in tokens, money, and time. OpenTelemetry traces, metrics, and logs in GreptimeDB, with seven Grafana dashboards.
dsh plugin add @tma1-ai/dsh-plugin-greptimedbBehavior-drift reports for DeepSeek Harness agents: compare two session logs and see exactly how behavior changed — tool sequence, file operations, reasoning volume, timing, retries — no third-party dependencies, CI-ready.
dsh plugin add dsh-driftwatchExplain failed DeepSeek Harness agent runs locally, then produce a redacted recovery plan. No model required.
dsh plugin add @huichangzz/dsh-postmortemReplay real DeepSeek Harness turns against Standard, Minimal, Anchored, or plugin candidates with frozen request-surface evidence
dsh plugin add @webwalkerhq/dsh-replay-labRequest Context Profiler for DeepSeek Harness — see what changed between model requests, and how cache reuse changed with it.
dsh plugin add dsh-context-lensLocal JSONL run traces for DeepSeek Harness sessions
dsh plugin add dsh-agent-run-logger