Me
Active days, streaks, a weekday × hour heatmap, projects, languages, commits, patches applied, models, reasoning tokens and your 5-hour and weekly limits.
What did you do with Codex?
Open your Codex history, or try the sample. Everything stays in your browser.
Runs in your browser. Nothing is uploaded.
Open your Codex history and see your rhythm, what you built, the tools and skills you leaned on, how close you came to your usage limits and how well it ran. Built on a data catalog of every number a usage dashboard can show.
Drop your sessions folder here
Codex keeps one file per session in ~/.codex/sessions. Drop the folder, or pick it. Add session_index.jsonl from ~/.codex for your threads' names.
Reading…
Capabilities
Every skill and who started it, every MCP server down to each tool, every built-in tool and subagent, with how often and how well.
Team
Weekly active people against seats, stickiness, retention cohorts, depth of use and adoption by team, with groups under five people hidden.
Telemetry
Turn duration and time to first token at p50 and p95, turn errors, command success, approvals, connector health, compactions and agent runs.
Catalog
Every metric with its definition, where Codex records it and its chart, the unified event schema, and the Codex OpenTelemetry names.
Reads Codex session files (.jsonl and the compressed .jsonl.zst) and events in the catalog's unified schema (.jsonl or .json). Prompt and response text is never read into the dashboard; thread names stay on your screen. Your auth.json and config are skipped even inside a dropped folder.