Codex
Blogizi for Codex
OpenAI's terminal/IDE coding agent. Run a second Codex session to draft and stage a blog post about what you're building — without leaving your coding workflow.
Blogizi drafts by default — publishing live always requires explicit confirmation.
Install
Add the skill, then ask Codex
SKILL.md discovery and CLI mechanics are covered on /agent — this is the Codex-shaped first command.
Terminal
npx skills add bytemindlab/blogizi-agent
export BLOGIZI_API_KEY=your_api_key
codex "draft a blog post about the auth changes in this repo and stage it on Blogizi"Full agent CLI docs: /agent
Parallel sessions
Code in one terminal. Draft in another.
blogizi-agent already grounds posts in the local repo’s code and git history, so a second Codex session can turn work you’re already doing into a draft without interrupting the first session.
Two sessions
# Terminal 1 — keep coding
$ codex "Refactor the auth module to use JWT tokens"
# Terminal 2 — draft a post about it, in parallel
$ codex "Draft a blog post explaining the JWT auth
refactor, targeting the keyword 'JWT authentication
Node.js', and stage it as a draft on Blogizi"
Codex (terminal 2): Reading git log and diff...
Drafted "Migrating to JWT Auth: What We Learned"
Pushed to Blogizi as a draft.
Review it in your dashboard — publishing live
is a separate step.Fit
Why Codex + Blogizi
- Draft directly from the repo you’re already working in — real code, real git history
- Run it in a second session without interrupting your coding session
- SKILL.md discovery — Codex learns the commands automatically
- Draft-first — Codex stages posts as drafts; publishing live stays a separate, explicit step
Keep coding. Stage the post in parallel.
One session builds the feature. Another drafts the story for Blogizi. You publish when you say so.
Prefer Claude? Claude Code works the same way — see /agent.