CLI integration
Codex CLI
Use the current built-in OpenTelemetry exporter and keep raw user prompt content redacted.
Open the Codex CLI guide Documentation overview
Connect your AI tools to Observagent.
Choose the path that matches what you run. Connect an existing CLI, desktop runtime, or local model API, then verify its telemetry in Observagent.
Choose a guide
Integration paths
Select the product-specific guide for the CLI, desktop runtime, or local model API you want to connect.
CLI integration
Claude Code
Export Claude Code logs, metrics, and beta traces directly through OTLP/HTTP.
Open the Claude Code guideCLI integration
Ollama
Route local Ollama requests through the Observagent proxy on Windows, macOS, or Linux.
Open the Ollama guideCLI integration
Antigravity
Send print-mode events or interactive session boundaries through a local Python adapter.
Open the Antigravity guideDesktop integration
ChatGPT desktop and Codex
Configure the local Codex runtime to export logs, metrics, and traces to your project.
Open the desktop guideEditor integration
Cursor
Stream Cursor Enterprise usage metrics and logs to Observagent over OpenTelemetry.
Open the Cursor guideIDE integration
GitHub Copilot
Send Copilot Chat, CLI, and desktop app telemetry through individual or org-managed OTLP export.
Open the GitHub Copilot guideAgent integration
Devin
Export Devin agent session and step telemetry to Observagent over OpenTelemetry.
Open the Devin guideAgent integration
Grok Bot
Stream Grok Bot activity through Cursor Enterprise's OpenTelemetry export, tagged cursor.surface=grok_bot.
Open the Grok Bot guideAgent integration
Grok Build
Export Grok Build's agent logs and metrics using its external OTEL variables or config.toml.
Open the Grok Build guideLocal API integration
LM Studio
Observe requests routed through an OpenAI-compatible local proxy without adding prompt content.
Open the LM Studio guideProduct FAQ
Frequently asked questions
What is Observagent?
Observagent is observability infrastructure for AI systems. It brings metrics, traces, logs, and usage data together so organizations can see how AI behaves across users, agents, applications, and departments.
How does Observagent work?
Your SDK, runtime, or local integration emits OpenTelemetry signals over HTTPS. Observagent organizes those signals for investigation and analysis to help teams improve performance, reliability, and cost.
What data gets sent to Observagent?
Integrations send OpenTelemetry traces, metrics, and logs over HTTPS. The exact fields depend on the SDK, runtime, and configuration you choose; Observagent does not require prompt or response text to accept telemetry.