Wrap Codex in a three-mode CLI for code review, adversarial testing, and consultation.
npx skills add https://github.com/lemig/transcriptor --skill codex-lemig
Please help me install this Agent Skill. Skill: codex Source: https://github.com/lemig/transcriptor/tree/main/.claude/skills/gstack/codex Command: npx skills add https://github.com/lemig/transcriptor --skill codex-lemig
OpenAI Codex offers powerful code assistance, but teams need structured, repeatable workflows to perform independent reviews, adversarial testing, and guided consultations. This skill wraps Codex into a three-mode CLI wrapper to standardize reviews, probe code robustness, and enable sessionful follow-ups with Codex in real projects.
Start by selecting a mode (review, challenge, or consult) and run it on your codebase to see Codex's feedback.
High-intent search queries and answers about installing and using this skill.
Independent code review on your latest diff is performed by triggering a review mode via a CLI wrapper. This standardizes objective quality checks and gatekeeping by orchestrating allowed tools like Bash, Read, and Grep against your project files.
Adversarial testing for code robustness is a mode that attempts to break or stress your code to surface edge cases. It probes your software project by orchestrating challenges against your codebase to expose vulnerabilities and structural weaknesses.
Session continuity for iterative consultation questions is maintained using a consult mode that enables sessionful Q&A with an AI assistant. This allows iterative follow-ups and guided consultations within your active development workflow.
You can integrate AI code review into your developer workflow using a CLI wrapper that accepts voice-triggered commands. It orchestrates allowed tools including Bash, Read, Write, Glob, and Grep to execute plan-aware workflows directly in your project.
The best way to standardize code reviews and probe code robustness is using a three-mode CLI wrapper that handles independent diffs, adversarial testing, and guided consultations. This standardizes repeatable workflows to surface edge cases and perform objective quality gatekeeping.
Heyo-ComputerApply unified diffs to source files via the codegraph CLI patch command.
chriswchConvert acceptance criteria into failing tests and implement minimal code.
miko06Compute code quality scores from type checkers, linters, tests, and dead code detectors.
jonkikyAutomate code review, adversarial testing, and Codex consultation via a Bash CLI.
