codex-issue-digest

Generate headline-first digests of openai/codex GitHub issues by label and time window.

Updated Aug 4, 2026
One-click install
npx skills add https://github.com/sjc786526-coder/RONDO --skill codex-issue-digest-sjc786526-coder
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: codex-issue-digest
Source: https://github.com/sjc786526-coder/RONDO/tree/main/multidev/.codex/skills/codex-issue-digest
Command: npx skills add https://github.com/sjc786526-coder/RONDO --skill codex-issue-digest-sjc786526-coder

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Maintainers of the openai/codex repository need a fast way to see which bug reports and enhancement requests are gaining user attention, without manually scanning dozens of GitHub issues across feature-area labels. ## Core Features & Use Cases - Label-Scoped Issue Collection: Runs a Python collector via the GitHub CLI to gather bug and enhancement issues filtered by owner labels such as tui, exec, or app, or across all labels. - Attention Scoring: Counts unique human interactions (new issues, comments, reactions) in a configurable window and marks elevated items with fire markers scaled to the window length. - Headline-First Reporting: Produces a summary-first digest with inline numbered issue links, optional details tables, and a source line citing script version, git head, and time window. - Use Case: A TUI owner asks for the digest of tui and exec issues over the past week and receives a one-line headline, the two hottest issues with links, and an optional details table. ## Quick Start Use $codex-issue-digest to run the Codex issue digest for labels tui and exec over the previous 24 hours.

Frequently Asked Questions about codex-issue-digest

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a GitHub issue digest for openai/codex by label?▼

Run the collector script with the labels and window you need, for example python3 collect_issue_digest.py --labels tui exec --window-hours 24. It uses the GitHub CLI to search issues labeled bug or enhancement plus your owner labels and outputs JSON for the digest.

How do I summarize GitHub issues across all labels or a custom time window?▼

Pass --all-labels to scan every feature-area label, and use --window with values like 48h, 7d, or past week, or --window-hours for a numeric lookback. You can also pin exact bounds with --since and --until ISO timestamps.

What do the fire emoji attention markers mean in the digest?▼

The markers reflect unique human users who opened, commented on, or reacted to an issue during the window. The baseline is 5 users for one fire and 10 for two over 24 hours, scaled linearly for longer or shorter windows, with bot activity excluded.

Does the issue digest require the GitHub CLI?▼

Yes, the collector shells out to the gh command for issue search, issue and comment hydration, and reactions endpoints. You need gh installed and authenticated against the target repository, which defaults to openai/codex but can be overridden with --repo.

Why might a reaction-only issue be missing from the digest?▼

GitHub issue search is seeded by updated_at, so an issue that only received new reactions without other activity may not appear in search results. Covering every reaction-only case would require a persisted snapshot store or a broader scan of labeled issues.