Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add gvastethecreator/codex-studio --skill codex-studio-setupgit clone --depth 1 https://github.com/gvastethecreator/codex-studioWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/gvastethecreator/codex-studio/codex-studio-setup)<a href="https://agentmods.dev/skills/gvastethecreator/codex-studio/codex-studio-setup"><img src="https://agentmods.dev/badge/skills/gvastethecreator/codex-studio/codex-studio-setup/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gvastethecreator/codex-studio/codex-studio-setup"><img src="https://agentmods.dev/badge/skills/gvastethecreator/codex-studio/codex-studio-setup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00026 | $0.01287 |
| Opus 5.5 | $0.00010 | $0.00515 |
| Sonnet 5.5 | $0.00005 | $0.00257 |
| Haiku 4.5 | $0.00003 | $0.00129 |
Grade A, and why
codex-studio-setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex Studio Setup
Use when a user asks to make this repo ready for first run, or when the welcome screen provides a setup handoff prompt.
Objective
Bring a Codex Studio checkout to a runnable local state:
- dependencies installed and repo scripts runnable
.env.localbootstrap present and sane- Studio Library initialized outside the repo
- SQLite migrations, default library, and default workspace created
- Codex CLI available, app-server-capable, and authenticated with ChatGPT login
codex app-serverreachable through the backend- UI and backend start cleanly
- closeout checks run once
Onboarding loop
Copy prompt and Ask Codex on the welcome surface use this skill as the Setup Prompt
(CODEX_STUDIO_SETUP_SKILL_PATH in lib/onboardingSetupPrompt.ts). Keep that
ownership. Do not move the prompt body into the renderer.
The product loop is detect, consent, mutate, stream, re-validate. One primary CTA:
- Missing Bun: open https://bun.sh/docs/installation. Never silent-install Bun.
- Missing Codex CLI: open https://github.com/openai/codex. Never silent-install Codex.
- ChatGPT login missing: Sign in from Studio Settings, or stop and ask the user to run
codex loginand choose ChatGPT. Login is not bundled. - Studio Library or Bootstrap Configuration missing: in-app Setup, or
bun run studio:onboard --setup, after explicit consent. - Everything else ready except Codex Product Runtime: start app-server through the local backend.
- Ready: Open Studio.
Ask Codex is an extra path when Codex CLI exists. Grok Imagine is an optional provider row, never a Studio installer.
Default Studio Library is a folder named Codex Studio in the user home. Existing
STUDIO_LIBRARY_DIR is kept. Do not auto-migrate AI-Studio-Library. Preferred Output
Path is not the generate destination. New generations go under outputs/<workspace>/
inside the Studio Library.
Safety
- Preserve dirty worktree changes you did not make.
- Never commit or print Provider Secret values.
- Do not store Provider Secrets in SQLite, catalog metadata, logs, screenshots, docs, or committed files.
- Do not delete, move, compact, or rewrite Studio Library data unless the user explicitly confirms that destructive operation.
- Keep generated images, SQLite files, transcripts, logs, local outputs, and
.env.localuncommitted. - Use
apply_patchfor manual repo edits. - Run broad gates only at closeout unless a focused failure requires a small targeted check.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · -5 lines 1e616f14dd3d
- 1mo ago First seen · 140 lines · 26 tokens per session scan A 40a2d34f83a4
codex-studio-setup is a skill published in the GitHub repository gvastethecreator/codex-studio (16 stars, last pushed 25d ago), licensed MIT. It adds 26 tokens to every session and 1,287 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
together-ai-inference
Serverless inference, fine-tuning, embeddings, image generation, and batch processing on 200+ open-source models via an OpenAI-compatible API. Use when you need fast, cost-effective access to open-source LLMs without managing infrastructure.
stable-diffusion-image-generation
State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.
crossgen-artist
Use CrossGen to plan and execute image generation, editing, inpainting, model selection, media-aware job monitoring, Gallery inspection, and asset export through MCP or its JSON CLI.
atet
Use Atet to create, generate, record, edit, and render visual media with a local CLI and coding agent. Trigger for AI image or video generation, screen, camera, microphone, or system-audio recording, non-destructive video editing, silence and filler-word removal, captions or subtitles, speaker framing, screen zooms…
query-kb
Load scoped repository context, then search and navigate a hraness/kb Markdown vault with hybrid text retrieval, exact metadata, bounded links and typed relationships, backlinks, whole-vault graph reports, and optional Git provenance. Use when an agent needs the applicable repository instructions, rationale, prior…
save-url-kb
Capture public or signed-in web content into a local Markdown knowledge base as an auditable source bundle with local assets. Use when the user asks to clip, save, scrape, or archive an article, social post or thread, GitHub or Discourse discussion, feed, inbox, private document, WhatsApp conversation, YouTube page…