The "observability" topic on GitHub groups 47 open-source projects in the DeepSeek Harness (DSH) ecosystem, led by dsh-maze with 84 GitHub stars. dsh-maze — DeepSeek Harness's execution maze—see how Agent actually works: maze timeline · data tracks · deterministic execution analysis · multi-session comparison | The execution maze for DSH agents: maze timeline, per-step data tracks, deterministic execution analysis, multi-session comparison. Every project here is indexed by DSH Universe with live GitHub data — stars, activity and install status — so you can compare and install directly.
MCP management console for the official DeepSeek Harness MCP client: /mcp command with health diagnostics and pipeline trial calls, a Settings MCP tab with server CRUD (approval-gated writes, automatic backups) and a tool trial console over the official tool pipeline (Apache-2.0, dsh-plugin).
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 is a DeepSeek Harness (DSH) observability plugin that reports GenAI trace data directly to Tencent Cloud Log Service (CLS). It observes DSH-native session, agent loop, LLM stream and tool lifecycles, converting them into the 5-layer span hierarchy model of the Tencent Cloud AI Agent observability spec (entry → agent → step → chat → tool), and reports directly to CLS via tencentcloud-cls-sdk-js, with no extra OTLP collector or sidecar deployment.
🐳 Whale Meter: how many tokens have you burned on DeepSeek Harness? Usage tiers, shareable achievement cards, accurate pricing for 46 models from 6 vendors (including China tiered pricing), before/after comparison of the 8·17 price change. Zero config, all data stays local | Token usage tiers & cost dashboard for DSH
Zero-touch runtime telemetry for DeepSeek Harness: every session writes its own report card (task/model/tools/failures/duration) into a human-readable journal. Pure-observer runtime fact telemetry.
洞察 — read-only insight panel for a DeepSeek Harness profile: where every plugin, service, tool and model came from, which config layer inserted or disabled it, and what is running right now.
DeepSeek Harness session replay and redacted sharing plugin: export real Agent trajectories as standalone interactive HTML for docs, demos and issue reports.
DSH Web UI file tracking plugin: records and views files the model read/wrote/edited, with line-numbered content and terminal-style line-by-line diff (red delete, green add, blue modify), hunk context folding, draggable height. Compatible with DSH dsh-v0.1.2-alpha.1, pure client-side, zero core changes.
DSH web plugin: see every session's conversation progress and token usage at a glance — all-windows overview, session analysis, cost estimation, and a token heatmap. Window stats plugin: see all sessions' conversation progress and Token usage at a glance — all-window overview, session analysis, cost estimation, Token heatmap.
🌊 Abyss Agency · Abyss — paints DeepSeek Harness multi-agent runs as an agency you can see: delegation lineage, per-agent cost and failure attribution, context level, attendance timeline; past cases can be replayed and exported as a Markdown retrospective in one click. A DeepSeek Harness plugin that turns an agent fleet into an office you can watch.
DSH Web UI file tracking plugin: records and views files the model read/wrote/edited, with line-numbered content and terminal-style line-by-line diff (red delete, green add, blue modify), hunk context folding, draggable height. Compatible with DSH dsh-v0.1.2-alpha.1, pure client-side, zero core changes.
Client-neutral AI-agent control plane: user global config overlay for Hermes/Codex/DSH/GitHub/Open Design/OpenHuman via one adapter contract (CC Switch observe-only). Modules: client-neutral-core (task/telemetry ledger, sidecar), services (orchestration/policy/receipts), read-only observer. Authority chain + exact-SHA CI gates.
Container-deployable agent harness on DeepSeek Harness — multi-provider routing, plugin-extensible personas and skills, local OpenTelemetry tracing and cost attribution.
Self-diagnosing plugin family for DeepSeek Harness (dsh): deterministic detour detection over the SessionEvent log, replay-verified optimization suggestions, and accepted-fix sediment.
DeepSeek Harness (dsh) plugin: live token cost of every chat, its subagents and the whole session tree — DeepSeek, OpenAI, Anthropic, Gemini, Kimi, Grok and Mistral prices from a bundled catalog, plus an append-only JSONL cost log in your project folder.
Processes tab for DeepSeek Harness: every agent and subagent of a conversation as a live console with paired tool requests and responses, installed as one profile bundle.
Money on the screen for the DeepSeek Harness web GUI: per-turn USD cost in the chat, a Cost tab with charts, ratings and an interactive spend timeline, and a local price table you control.