Trace Gemini CLI sessions, agent steps, model calls, and tool invocations in Arize AX for full observability.Gemini CLI is Google’s open-source command-line agent for Gemini. The Arize Coding Harness Tracing instruments Gemini CLI sessions and exports OpenInference spans to Arize AX.
Launch Arize AX
To get started, sign up for a free Arize AX account and get your Space ID and API Key:- Log in at app.arize.com
- Click Settings and copy the Space ID
- Open the API Keys tab and create or copy an API key
Install
Curl installer
macOS / Linux:Local clone
~/.arize/harness/config.json, and registers the hooks in ~/.gemini/settings.json.
Configuration
Credentials live in~/.arize/harness/config.json. Environment variables override values in config.json and can be set in your shell profile.
Redaction controls
EachARIZE_LOG_* flag accepts "true" or "false" and defaults to "true". Set to "false" to opt out per category:
| Flag | Redacts |
|---|---|
ARIZE_LOG_PROMPTS | User prompt and assistant response text |
ARIZE_LOG_TOOL_DETAILS | Tool names and arguments |
ARIZE_LOG_TOOL_CONTENT | Tool call output content |
Observe
Once tracing is enabled, Gemini CLI activity is streamed to Arize AX. You’ll see:- Session traces — each session bracketed by
SessionStartandSessionEnd - Agent spans — paired
BeforeAgent/AfterAgentevents for each agent step - LLM spans — paired
BeforeModel/AfterModelevents with model info and tokens - Tool spans — paired
BeforeTool/AfterToolevents with input, output, and duration - Session grouping — all spans from the same session grouped by
session_id
Hooks Captured
| Hook | Span Kind | Description |
|---|---|---|
SessionStart | CHAIN | Session initialized |
BeforeAgent | CHAIN | Agent step started, records input |
AfterAgent | CHAIN | Agent step completed, records output |
BeforeModel | LLM | Model call started, records prompt and model |
AfterModel | LLM | Model call completed, records output and tokens |
BeforeTool | TOOL | Tool invocation started, records name and input |
AfterTool | TOOL | Tool invocation completed, records output and duration |
SessionEnd | CHAIN | Session teardown, state file cleanup |