OpenTelemetry

MCP observability turns the Model Context Protocol into a two-way interface: AI agents don’t just call tools, they receive ground-truth telemetry from kernel tracepoints, eBPF programs, and kprobes. By exposing low-level system instrumentation through MCP servers, agents get a real-world model of the live system instead of hallucinated state — closing the observability gap that traditional APM leaves wide open. What Is MCP and Why It Needs Observability The Model Context Protocol (MCP) is an open standard that standardizes how AI agents discover and invoke tools, resources, and prompts. Instead of every agent building bespoke integrations with every service, MCP defines a common protocol: a host (the agent runtime) connects to MCP servers, which expose tools the model can call and resources it can read. ...

DeepSeek harnesses — the scripts, agents, and pipelines that call DeepSeek models — become production systems the moment they leave your laptop, and production systems need observability. The fastest way to get it is to instrument your harness with OpenTelemetry and export the traces to Langfuse, an open-source AI engineering platform that turns raw OTLP spans into searchable, debuggable agent traces. This guide walks you through the exact setup, from the OTLP endpoint and authentication to the OpenAI SDK integration and self-hosting, so you can trace every DeepSeek call end to end. ...

OpenTelemetry tracing for DeepSeek Harness lets you export every agent session, LLM call, and tool invocation as a standard OTLP trace tree to backends like Jaeger, Grafana Tempo, SigNoz, or Langfuse. You add it by installing a community plugin that implements the official @deepseek-ai/dsh-session-telemetry seam, configure an OTLP endpoint and a privacy mode, and then read the GenAI trace tree to debug agent loops, retries, and token usage. What is DeepSeek Harness and why it needs tracing DeepSeek Harness is the official open-source agent framework from DeepSeek, written in TypeScript with the tagline “Everything is a Plugin.” Its official repository has roughly 123,000 GitHub stars, making it one of the most popular agent harnesses in the ecosystem. The framework orchestrates multi-step agent loops: it plans, calls LLMs, invokes tools, spawns subagents, and retries failed steps. Each of those steps is a potential failure point, and without tracing you are effectively debugging a black box. ...

Multi-agent workflow observability means capturing every delegation hop, tool call, and sub-agent handoff as first-class telemetry instead of relying on flat log lines. Because LLM agents fail silently, teams must trace intermediate reasoning, run offline evals against synthetic datasets, and add regression suites before shipping. This guide explains the observability gap, how to trace delegation hops, and how to build a practical observability and testing stack in 2026. Why Multi-Agent Delegation Demands a New Observability Mindset A multi-agent system distributes a complex goal across specialized agents that hand work to one another. The rationale is straightforward: multi-agent systems solve problems that are difficult or impossible for a single monolithic agent, which is the core justification for delegation patterns in the first place. When an agent delegates a subtask to a colleague agent, a supervisor, or a sub-process, the resulting behavior is emergent, non-deterministic, and often invisible to the humans who wrote the system. ...

OpenTelemetry is the standard way to add structured tracing, metrics, and logs to AI agents in 2026 — covering token usage, tool call latency, and multi-agent context propagation with a single SDK and vendor-neutral backends. Why Traditional Observability Fails for AI Agents Traditional APM tools like Datadog APM or New Relic were designed for deterministic request/response cycles: a user hits an endpoint, a function runs, a database query fires, a response returns. The execution path is fixed, latency is bounded, and errors are binary. AI agents break every one of these assumptions. An agent reasoning chain is non-deterministic — the same input prompt can trigger three tool calls in one run and seven in the next. Execution duration ranges from 500ms for a fast LLM call to 3+ minutes for a multi-step agent that searches the web, queries a database, and synthesizes results. Without agent-native spans, you cannot tell which tool call caused a timeout or why a particular run cost $0.40 while a similar one cost $0.03. Traditional APM measures function latency in microseconds and ignores tokens entirely. The LLM observability platform market recognized this gap — growing to an estimated $2.69 billion in 2026 and projected to reach $9.26 billion by 2030 at a 36.2% CAGR. OpenTelemetry’s GenAI Semantic Conventions fill that gap with a purpose-built span model for LLM operations, agent reasoning loops, and tool executions that traditional APM never anticipated. ...

Arize Phoenix is a free, open-source LLM observability platform that gives developers full-stack visibility into LLM applications — tracing requests, evaluating outputs, and debugging RAG pipelines — without requiring a cloud subscription or vendor account. It runs locally in a Python process or scales to Docker and Kubernetes for production deployments. What Is Arize Phoenix and Why It Matters in 2026 Arize Phoenix is an open-source observability platform built specifically for LLM applications, agents, and retrieval-augmented generation (RAG) pipelines. Unlike generic APM tools, Phoenix understands LLM-native concepts — spans, traces, embeddings, prompts, retrieved contexts, and model outputs — and surfaces them in a UI designed for AI engineers. As of 2026, Phoenix has surpassed 9,000 GitHub stars, making it one of the most-adopted open-source observability tools in the AI ecosystem. The platform is backed by Arize AI but released under a permissive open-source license, meaning you can run it entirely on your own infrastructure with no usage caps or feature gating. ...