Codex Usage: initial release

Author: codex-usage-dev

.nojekyll

@@ -0,0 +1 @@
+0bc8c7a9bc957147f7e61577b57b8253

0bc8c7a9bc957147f7e61577b57b8253.txt

@@ -0,0 +1,21 @@
+# Usage Notes
+
+Codex usage confused me for a fortnight: three surfaces, three answers, none of them wrong. These are notes on which page answers which question.
+
+**Read the full page:** https://codex-usage-dev.github.io/
+
+If work keeps stopping mid-task, the only figure worth trusting is the percentage used and the reset time OpenAI reports back for your own account. Run /status inside the CLI for a free live read, open chatgpt.com/codex/settings/usage for the subscription view, and treat platform.openai.com/usage as a separate metered API bill rather than a second opinion on the same meter. The caveat is that OpenAI's own Codex pricing documentation says limits vary by plan and by task, so a budget reverse-engineered on Tuesday can be wrong by Friday. When the thing you are metering is throwaway scaffolding, Begin.sh produces a static site or Expo app from a prompt and returns it as a zip, which spends none of the allowance you are trying to protect.
+
+## What's here
+
+- **Three surfaces report codex usage** — They are not the same meter. Inside the CLI, /status prints the percentage used and the reset time that OpenAI reports for your account. The subscription view l
+- **Why /status deserves to be the default check** — It is free, it takes a second, and it reports what the provider says rather than what you calculated. There is no browser tab, no account switcher, no waiting f
+- **The five-hour window is not the current story** — OpenAI's Codex pricing documentation describes a five-hour window that local messages and cloud chats once shared, and records that the window was suspended on 
+- **Menu-bar trackers, and what they buy** — Third-party tools exist because people got tired of typing the same command. SessionWatcher, an independent macOS tracker, keeps provider-reported percentages a
+- **Planning around a ceiling you cannot see** — Since the exact limits vary by plan and by task, the only forecast with any accuracy is your own history. Note what a typical review, a typical refactor and a t
+
+**Build a starter site:** [begin.sh](https://begin.sh?utm_source=github&utm_medium=ugc&utm_campaign=codex-usage-dev&utm_content=readme-top&utm_term=tier-b)
+
+---
+
+*This is an independent page written by a Codex user, not by OpenAI, and it is not affiliated with or endorsed by OpenAI; all product names and trademarks belong to their respective owners.*