Open Agent Spec

Open Agent Spec is Oracle’s portable, platform-agnostic configuration language for describing agents and agentic systems. Its tracing extension standardizes how agent and flow executions emit traces across runtime adapters. Arize AX captures every Agent Spec run — agent invocations, LLM turns, and tool calls — via the openinference-instrumentation-agentspec package.

Prerequisites

Python 3.10+
An Arize AX account (sign up)
An OPENAI_API_KEY from the OpenAI Platform (the example loads the Agent Spec agent through the LangGraph adapter, which calls OpenAI)

Launch Arize AX

Sign in to your Arize AX account.
From Space Settings, copy your Space ID and API Key. You will set them as ARIZE_SPACE_ID and ARIZE_API_KEY below.

Install

The langgraph extra pulls the LangGraph adapter (and langchain-openai) that turns an Agent Spec definition into a runnable agent:

pip install arize-otel \
  openinference-instrumentation-agentspec \
  "pyagentspec[langgraph]"

Configure credentials

export ARIZE_SPACE_ID="<your-space-id>"
export ARIZE_API_KEY="<your-api-key>"
export ARIZE_PROJECT_NAME="agentspec-tracing-example"
export OPENAI_API_KEY="<your-openai-api-key>"

Setup tracing

# instrumentation.py
import os

from arize.otel import register
from openinference.instrumentation.agentspec import AgentSpecInstrumentor

tracer_provider = register(
    space_id=os.environ["ARIZE_SPACE_ID"],
    api_key=os.environ["ARIZE_API_KEY"],
    project_name=os.environ["ARIZE_PROJECT_NAME"],
)

AgentSpecInstrumentor().instrument(tracer_provider=tracer_provider)
print("Arize AX tracing initialized for Open Agent Spec.")

Run Open Agent Spec

# example.py

# Importing instrumentation first ensures tracing is set up before the
# Agent Spec agent runs.
from instrumentation import tracer_provider

from pyagentspec.adapters.langgraph import AgentSpecLoader
from pyagentspec.agent import Agent
from pyagentspec.llms import OpenAiConfig

# Describe the agent in Agent Spec, then load it into a runnable
# LangGraph agent. The LLM reads OPENAI_API_KEY from the environment.
agent = Agent(
    name="assistant",
    description="A general-purpose assistant without tools",
    llm_config=OpenAiConfig(
        name="openai-gpt-5.4-mini",
        model_id="gpt-5.4-mini",
    ),
    system_prompt="You are a helpful assistant. Answer politely.",
)

langgraph_agent = AgentSpecLoader().load_component(agent)

response = langgraph_agent.invoke(
    input={
        "messages": [
            {
                "role": "user",
                "content": (
                    "Why is the ocean salty? Answer in two sentences."
                ),
            }
        ]
    },
    config={"configurable": {"thread_id": "1"}},
)

print(response["messages"][-1].content.strip())

Expected output

Arize AX tracing initialized for Open Agent Spec.
The ocean is salty because rivers continuously dissolve mineral salts from rocks and soil and carry them to the sea, where they accumulate over millions of years. Water leaves the ocean through evaporation, but the dissolved salts remain and steadily concentrate over time.

Verify in Arize AX

Open your Arize AX space and select project agentspec-tracing-example.
You should see a new trace within ~30 seconds with an AgentExecution[assistant] AGENT span carrying the prompt as input and the agent’s answer as output, wrapping an LlmGenerationSpan LLM child span with the model call and token usage. Any tools the agent invokes appear as child TOOL spans with their inputs and outputs.
If no traces appear, see Troubleshooting.

Check from the skill, CLI, or SDK

Confirm spans are actually reaching your Arize AX project. Use whichever fits your workflow — the skill and CLI work for any framework; the SDK check is shown for each language.

Arize skill (agent)
AX CLI
SDK

Install the Arize Skills plugin and let your coding agent check for you:

npx skills add Arize-ai/arize-skills

Then prompt your agent:

Use the arize-trace skill to export and analyze recent traces from my project. Confirm spans are arriving, and summarize any errors or latency issues.

Export recent spans for your project — any rows mean traces are landing:

ax spans export "$ARIZE_PROJECT_NAME" --space "$ARIZE_SPACE_ID" \
  --limit 5 --stdout | jq 'length'

A non-zero count confirms spans reached Arize AX. Run ax auth login first if you have not authenticated. See the ax spans reference.

Query the project’s spans and check that at least one came back.

import os
from arize import ArizeClient

client = ArizeClient(api_key=os.environ["ARIZE_API_KEY"])
resp = client.spans.list(
    project=os.environ["ARIZE_PROJECT_NAME"],
    space=os.environ["ARIZE_SPACE_ID"],
    limit=5,
)
count = len(resp.spans)
print(
    f"{count} span(s) found" if count else "No spans yet — recheck setup"
)

// Reads ARIZE_API_KEY from the environment.
import { listSpans } from "@arizeai/ax-client";

const { data: spans } = await listSpans({
  project: process.env.ARIZE_PROJECT_NAME!,
  space: process.env.ARIZE_SPACE_ID!,
  limit: 5,
});
const count = spans.length;
console.log(
  count ? `${count} span(s) found` : "No spans yet — recheck setup",
);

client, err := arize.NewClient(
    arize.Config{APIKey: os.Getenv("ARIZE_API_KEY")},
)
if err != nil {
    log.Fatal(err)
}
resp, err := client.Spans.List(ctx, spans.ListRequest{
    Project: os.Getenv("ARIZE_PROJECT_NAME"),
    Space:   os.Getenv("ARIZE_SPACE_ID"),
    Limit:   5,
})
if err != nil {
    log.Fatal(err)
}
fmt.Printf("%d span(s) found\n", len(resp.Spans))

SDK span references: Python · TypeScript · Go.

Troubleshooting

No traces in Arize AX. Confirm ARIZE_SPACE_ID and ARIZE_API_KEY are set in the same shell that runs example.py. Enable OpenTelemetry debug logs with export OTEL_LOG_LEVEL=debug and re-run.
Agent spans missing. AgentSpecInstrumentor().instrument(...) must run before the agent is loaded and invoked. Make sure instrumentation.py is the first import in your entry point.
ModuleNotFoundError: No module named 'langgraph'. Install the LangGraph adapter with the langgraph extra: pip install "pyagentspec[langgraph]". Other runtime adapters (agent-framework, autogen, crewai) have their own extras.
401 from OpenAI. Verify OPENAI_API_KEY is set and has access to the model in the example. Swap gpt-5.4-mini for a model your key can call.

Prerequisites

Launch Arize AX

Install

Configure credentials

Setup tracing

Run Open Agent Spec

Expected output

Verify in Arize AX

Check from the skill, CLI, or SDK

Troubleshooting

Resources

Open Agent Spec Documentation

Agent Spec Tracing

OpenInference Agent Spec Instrumentor

​Prerequisites

​Launch Arize AX

​Install

​Configure credentials

​Setup tracing

​Run Open Agent Spec

​Expected output

​Verify in Arize AX

​Check from the skill, CLI, or SDK

​Troubleshooting

​Resources

Open Agent Spec Documentation

Agent Spec Tracing

OpenInference Agent Spec Instrumentor

Prerequisites

Launch Arize AX

Install

Configure credentials

Setup tracing

Run Open Agent Spec

Expected output

Verify in Arize AX

Check from the skill, CLI, or SDK

Troubleshooting

Resources