docs: add GPT-5.3 OpenCode delivery playbook#12
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: c8b179bb30
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| | new task request with ambiguity | `Prompt Wizard`, `workflow` | clear task statement with in-scope/out-of-scope and verification checklist | | ||
| | editing unfamiliar code | `code-search`, `Pattern Analysis` | file map, current conventions, and safe insertion points | | ||
| | implementing code changes | `language-conventions`, `workflow` | minimal-diff changes aligned to existing standards | | ||
| | adding or changing tests | `Testing Strategy` | test pyramid plan and targeted coverage additions | | ||
| | prompt/model behavior design | `Prompt Engineering` | versioned prompt contract (role, constraints, schema, examples) | |
There was a problem hiding this comment.
This table maps workflows to skill names that are not shipped in this repo (for example Prompt Wizard, Pattern Analysis, Testing Strategy, Prompt Engineering, and AI Evaluation), so readers following the “default runbook” cannot execute those steps. The currently available skills are enumerated in skills/README.md and the skills/*/SKILL.md directories, and these labels do not match, which makes the playbook misleading at the point of use.
Useful? React with 👍 / 👎.
1 participant
Summary
Testing