Project Rosetta Stone

Arize AX is an AI engineering platform from Arize AI for observing and evaluating AI agents, including voice agents. It provides a unified session and trace view that combines audio, transcripts, latency metrics, tool calls, and evaluation results, allowing failures such as dead air, interruptions, transcription errors, and incorrect tool actions to be investigated together. The platform supports tracing real-time audio with OpenInference semantic conventions, audio-native evaluations for factors such as sentiment, latency, interruptions, and task success, and agent experiments that test fixes against failed traces. It is available through the Arize AX product interface and a terminal-based setup.

Phoenix is an open-source AI observability platform developed by Arize for experimenting with, evaluating, and troubleshooting LLM applications and agents. It traces application runtime with OpenTelemetry-based instrumentation, supports response and retrieval evaluations, and provides versioned datasets, experiment tracking, prompt management, and a playground for comparing models, adjusting parameters, and replaying traced calls. Phoenix also includes PXI, an AI engineering agent for debugging traces and iterating on prompts, plus a remote MCP server for querying traces, datasets, and experiments from MCP clients. It can run locally, in containers, through Kubernetes and Helm, or in the cloud, and is distributed through the `arize-phoenix` Python package, Conda, Docker images, and related TypeScript and Python packages.

Arize Skills is a set of installable skills for AI coding agents, maintained by Arize, that guides agents through adding observability, creating datasets, running experiments and evaluations, and optimizing prompts for LLM applications. The skills use the ax CLI to interact with Arize and include workflows for tracing and debugging applications, OpenInference-based instrumentation, dataset management, experiments, LLM-as-judge evaluators, annotations, prompt optimization, provider credentials, and production monitoring. They can be installed into agents such as Cursor, Claude Code, Codex, GitHub Copilot, and Windsurf through npx, the repository's installers, the ax CLI, or a Claude Code plugin. The repository is licensed under MIT.

AgentOps is an open-source observability and developer-tools platform for building, evaluating, and monitoring AI agents. Its Python SDK records agent sessions and LLM calls for dashboard-based session replays, step-by-step execution graphs, debugging, analytics, cost tracking, and benchmarking. Developers instrument code with decorators for sessions, agents, operations, tasks, and workflows; these support nested span hierarchies, input and output recording, exception handling, asynchronous functions, generators, and custom attributes. The SDK provides integrations with agent frameworks and LLM systems including CrewAI, AG2/AutoGen, Agno, LangGraph, and the OpenAI Agents SDK, and the platform can be self-hosted with its dashboard and API backend. The app is released under the MIT license.

Multi-Agent Workbench is a local-first browser control room for orchestrating coding agents such as Claude Code, Codex CLI, and generic PTY-based command-line agents. Agents work in task rooms under editable YAML role cards that define permissions, speaking style, and decision boundaries; a rule-driven orchestrator pauses execution and risky commands for human approval, while persistent sessions expose terminal activity and status. The system uses a local event-driven architecture with a React interface, Fastify backend, REST API, server-sent events, agent adapters, an approval service, and a SQLite event store. Every action passes through the API and is recorded in an append-only event log, whose events are projected into timelines, decisions, terminal views, artifacts, and diff views. It also provides task templates, Markdown reports and decision logs, real-time room updates, and configurable risk resolutions. Code-writing agents are intended to use isolated Git worktrees, with diff and patch review before changes reach the main workspace; the repository marks this worktree and merge flow as in progress.