Trace every AI agent execution, step by step
See every LLM call, tool execution, retrieval, and database query as connected spans in a detailed execution timeline..
Trace every AI agent execution, step by step: Docs (opens in new tab)Observability
Every agent decision, captured.
Not just another log.
Add the SDK and every reasoning step, tool call, retrieval and handoff your agent makes in production becomes a structured, replayable trace.
Trusted by teams building and deploying AI agents
What is AI Agent Observability?
A structured record of every step your agent takes. Netra captures each LLM call, tool call and retrieval as a span inside a trace, with inputs, outputs, latency, tokens and cost. Sessions, users and tenants group those traces, so you can go from a customer's complaint to the exact step that caused it.
Find where your AI agent failed, not just which request failed
Trace every LLM call, tool execution, and retrieval as an individual span, with inputs, outputs, latency, and token usage. Pinpoint the exact step behind an incorrect response or failed agent execution.
See the metadata of every step
Inspect span-level inputs, outputs, model parameters, token counts, and execution details to understand how your AI agent behaves..
See the metadata of every step: Docs (opens in new tab)
Monitor AI agent cost, latency, and token usage
Identify slow or expensive execution steps and track token usage, latency, and cost across models, tenants, users, and time ranges..
Monitor AI agent cost, latency, and token usage: Docs (opens in new tab)
Tell every step apart by type
Span, generation, embedding, tool and agent types are assigned automatically, so each step shows up as what it is.
Tell every step apart by type: Docs (opens in new tab)Track AI agent costs for every customer
Attribute AI agent usage, token consumption, and costs to individual customers using tenant-level tracing. Monitor spending and performance across agents, sessions, and users from the same platform.
Switch between agent, tenant, session and user views
Agent cost, latency and tool use; tenant spend; sessions as whole conversations; every trace for one user.
Switch between agent, tenant, session and user views: Docs (opens in new tab)Attribute cost and usage to each tenant
Sessions, traces and total cost for every customer, sorted by spend, so the heaviest users are one glance away.
Attribute cost and usage to each tenant: Docs (opens in new tab)Debug the journey, not 40 requests
Group multi-turn interactions by session and user, then drill from session totals to the exact span.
Debug the journey, not 40 requests: Docs (opens in new tab)Save the views your team checks
Search by name, time range, status or any custom attribute, choose your columns, and save the filter.
Save the views your team checks: Docs (opens in new tab)Monitor AI agent performance across your entire system
Track AI agent performance with real-time visibility into latency percentiles, error rates, token usage, and cost. Build custom dashboards to monitor agent activity across your production environment.
See detailed analytics of your traces
Monitor latency percentiles, error rates, request volume, token usage, cost, and time to first token across your AI agents.
See detailed analytics of your traces: Docs (opens in new tab)Build custom dashboards
Six chart types over the same metrics, segmented by environment, user, session or tenant.
Build custom dashboards: Docs (opens in new tab)Filter down to what matters
Narrow by model, user, session, error or PII flags, and by latency or cost range.
Filter down to what matters: Docs (opens in new tab)Reuse widgets across dashboards
Add a widget to as many dashboards as you like, or clone a whole one, so every team reports on the same numbers.
Reuse widgets across dashboards: Docs (opens in new tab)Debug AI agents in your IDE, and prevent repeat failures
Investigate failing AI agent traces directly in your IDE through Netra's MCP server. Identify the issue, turn the production failure into a test case, and catch the same regression in future evaluations.
Debug AI agents using MCP
Bring traces, spans and sessions into Cursor, Claude Code, Windsurf or Codex through the Netra MCP server.
Debug AI agents using MCP: Docs (opens in new tab)Look up any trace or session
Fetch spans by trace ID, filter traces with pagination, and inspect session metrics without leaving your editor.
Look up any trace or session: Docs (opens in new tab)Add a trace as a test case
Turn a failing production trace into a test case, so the bug you just found becomes a check in your next evaluation.
Add a trace as a test case: Docs (opens in new tab)Grow a dataset from your editor
Create a dataset and add items one at a time or in bulk, straight from your coding agent.
Grow a dataset from your editor: Docs (opens in new tab)OpenTelemetry-native observability for your AI agents
Built on OpenTelemetry, Netra brings AI agent tracing into your existing stack. Capture model calls, tool execution, and agent behavior with minimal instrumentation, while keeping the observability tools and infrastructure you already use.
- OpenTelemetry
- LangChain
- LangGraph
- CrewAI
- LlamaIndex
- OpenAI
- Anthropic
- LiveKit
- Vapi
- Pinecone
- Datadog
- MCP
Instrument once, in minutes
One Netra.init() traces 30+ AI libraries, including model providers, frameworks and vector databases, with no code changes.
Add decorators and manual spans
@workflow, @agent, @task and @span for minimal-code tracing, plus manual spans for anything custom.
Trace voice agents end to end
Trace voice agents end-to-end, from speech-to-text and LLM reasoning to text-to-speech. Identify latency at each stage of the voice interaction within a single execution trace.
Docs (opens in new tab)Control what's captured
Choose which libraries are instrumented, and switch off prompt and response capture while keeping latency, token and cost metrics.
Docs (opens in new tab)In production
Every agent, every tenant, in real time
How Pencil watches every agent, every tenant
Netra turned our ad-agent stack from a black box into a clear control center. We can see what every agent is doing, for every tenant, in real time, and Netra now processes over 1M spans every day for Pencil. Issues that once took hours to debug now take minutes.
Frequently Asked Questions
Everything You Need to Know About AI Agent Observability
Can't find the answer here? The Observability docs go deeper, or talk to our team.
What is AI agent observability?
AI agent observability is the ability to trace, monitor, and understand every step an AI agent takes. Netra captures each LLM call, tool call, and retrieval as a span within an execution trace, including inputs, outputs, latency, token usage, and cost. By connecting traces across sessions, users, and tenants, teams can identify failures, debug agent behavior, and monitor performance in production.
How is AI agent observability different from traditional application monitoring?
Traditional application monitoring focuses on metrics such as request latency, error rates, and infrastructure performance. AI agent observability also captures the steps behind an agent's response, including model calls, tool execution, retrievals, and multi-turn interactions. This helps teams investigate behavioral failures even when requests complete successfully.
How does Netra help debug AI agents in production?
Netra captures structured traces of AI agent execution, allowing teams to inspect individual spans, identify failed operations, and analyze latency, token usage, and cost. Developers can also retrieve traces through Netra's MCP server and turn production failures into test cases for future evaluations.
What does an AI agent trace in Netra include?
A Netra trace captures the complete execution of an AI agent request, including LLM calls, prompts, model responses, tool invocations, retrieval steps, token usage, latency, cost, errors, and custom metadata. Each step appears as a span, helping teams understand how the agent executed a task and where failures occurred.
What is the difference between a trace and a span in AI agent observability?
A trace represents the complete execution of an AI agent request or workflow. A span represents an individual operation within that trace, such as an LLM call, tool execution, or retrieval. Together, traces and spans help teams follow agent execution and investigate issues at each step.
How can I instrument my AI agent with Netra?
Netra supports instrumentation through its SDK, automatic tracing of supported AI libraries, decorators, and manually created spans. Teams can capture agent execution with minimal code changes and add custom instrumentation when needed.
How do I track agents, sessions and users?
Use the @agent decorator for agent execution, set_session_id to group related interactions, and set_user_id to attribute traces to a person. set_tenant_id does the same for a customer.
How are costs tracked?
Token usage and cost are captured automatically for supported providers and can be broken down by model, tenant, user or time range. For a custom gateway or unsupported provider, attach usage as span attributes.
Can I keep prompts and responses out of Netra?
Yes. Content capture can be switched off while latency, token and cost metrics are kept, and you choose exactly which libraries are instrumented.
Start today
See the step that broke before the ticket lands
Install the SDK and see your first trace in minutes. Free to start.