Tool · Models & evaluation

Usage Capture

Read visible usage values with separate Codex and Ollama adapters and preserve uncertain readings.

Recorded synthetic example. Source with a complete offline example.

See the idea in action.

five_hour=73, weekly=41, strategy=labels; no account or desktop needed.

Example input

5h: 73%; Weekly: 41% (synthetic labeled OCR).

Captured result

{
  "five_hour": 73.0,
  "weekly": 41.0,
  "strategy": "labels"
}

From the repository root, follow the dependency requirements in the README. This example uses supplied synthetic material.

python -m pip install -e ./codex
codex-usage parse-text codex/examples/synthetic-ocr.txt

five_hour=73, weekly=41, strategy=labels; no account or desktop needed.

OCR output can be useful evidence if capture, parse strategy and uncertainty are kept together.

Related adapters packaged together

Two separately authored Windows usage-display scripts, for Codex and Ollama.

Keeps distinct installable adapters and original APIs together around one visitor job.

They are peer adapters with different screen assumptions, not one derived from the other.