codex-usage-tracking is an open-source CLI tool and dashboard for developers to investigate aggregate token usage, costs, caching, and thread patterns when working with OpenAI Codex APIs. Below are 7 llm eval & observability apps with similar functionality to codex-usage-tracking, matched by what each product actually does — not ranked or scored. Explore each to find the closest fit for your use case.
codex-stats is an open-source CLI tool that provides local observability and analytics for Codex AI coding sessions. It tracks usage data and stores analytics locally, helping developers monitor and analyze their AI-assisted coding workflows.
codex-quota-monitor is a command-line and self-hosted tool for monitoring Codex OAuth quota and usage, providing routing scores, Prometheus metrics, and a NixOS module. It is designed for system administrators managing API usage.
Codex CLI usage monitor — fetches rate limits from OpenAI's ChatGPT backend API
codex-limit is an open-source macOS menu bar widget that allows users to track their Codex weekly and 5-hour usage limits. It provides real-time quota monitoring for efficient resource management.
codexcomp is an open-source CLI tool that acts as a local proxy for the OpenAI Codex CLI. It allows developers to handle responses locally, apply reasoning truncation, and maintain compatibility with the official OpenAI API. The tool is designed for developers working with AI code generation workflows.
codex-bell is an open-source CLI tool that alerts developers with a beep when OpenAI Codex finishes a task or needs user input. It helps streamline workflows by providing timely notifications, making it easier for developers to manage long-running Codex operations.
Codex Enhanced is a fork of OpenAI Codex designed with an operator-first approach, emphasizing visible automation and direct local control for coding workflows. It addresses the needs of users who require sharper session management, account switching, and workflow boundaries, consolidating these operations into a unified terminal user interface (TUI) rather than dispersing them across scripts. The tool integrates features such as loop automation, clawbot session binding, and rapid session respawn, all aimed at improving release velocity and operational clarity. A key aspect of Codex Enhanced is its support for loop automation, offering before-turn and after-turn runners that make queue states and progress visible within the TUI. This transparency allows operators to track origin markers, queue status, and stop behaviors directly, rather than relying on hidden background processes. The platform also incorporates a Feishu clawbot, enabling workspace-backed session discovery and binding, caching unread messages before binding, and forwarding assistant responses back into active sessions. Origin markers in chat transcripts help maintain legibility by labeling Feishu-originated turns and loop replies, even when automation and remote chat entry points are mixed. Codex Enhanced delivers multiple loop context modes—embed, ephemeral, and persistent—each tailored for different automation scenarios. The embed mode submits prompts directly into the main thread, ephemeral mode processes prompts in a private hidden thread and discards the context after completion, and persistent mode retains a private rollout that refreshes with the latest messages for each run. The tool supports fast self-restart and session resumption through commands like '/respawn', 'Ctrl-X R', and 'SIGUSR1', preserving the current session and enabling uninterrupted workflows. Installation is available via PyPI runtime wheels or by building the CLI locally, with each wheel including a platform-specific native codex binary. Codex Enhanced is distributed as a CLI tool and is geared toward operators and developers seeking greater control and visibility in their coding agent workflows.
