miuuyy/codex-chatgpt-web is a launcher that lets Codex use ChatGPT Web models, including account-based Pro models, as native choices for coding tasks. It is for Codex users who want ChatGPT Web conversations, images, streaming, and task tools without using their Work or Codex quota.
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 miuuyy/codex-chatgpt-web --skill pro-researchergit clone --depth 1 https://github.com/miuuyy/codex-chatgpt-webWrote 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/miuuyy/codex-chatgpt-web/pro-researcher)<a href="https://agentmods.dev/skills/miuuyy/codex-chatgpt-web/pro-researcher"><img src="https://agentmods.dev/badge/skills/miuuyy/codex-chatgpt-web/pro-researcher/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/miuuyy/codex-chatgpt-web/pro-researcher"><img src="https://agentmods.dev/badge/skills/miuuyy/codex-chatgpt-web/pro-researcher.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 | $0.00051 | $0.01177 |
| Opus 5.5 | $0.00020 | $0.00471 |
| Sonnet 5.5 | $0.00010 | $0.00235 |
| Haiku 4.5 | $0.00005 | $0.00118 |
Grade A, and why
pro-researcher 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 yesterday.
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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pro Researcher
Use ChatGPT 6 Pro as a research partner. These agents can work with a large context, inspect a repository, and investigate web sources through the tools available to them. Their strength is deep investigation: finding architectural problems, connecting research to the code, challenging assumptions, and helping choose a direction.
Use them proactively when that broader view would help. You do not need a precise bug or a finished plan before asking; an abstract topic or an uncertain architectural direction is a valid assignment. Handle routine implementation locally or with coding agents. Pro researchers can edit local files, but research is their primary role.
Select the researcher
- Inspect the available sub-agent tool and explicitly select ChatGPT 6 Pro. In Codex Web GPT, use
model: "chatgpt-web/gpt-6-pro",reasoning_effort: "max", andfork_context: false. Use the general-purpose agent role so a coding-only role does not narrow the investigation. - Start with no inherited conversation history. Give the researcher a self-contained assignment instead. Fork history only if the user explicitly asks for it.
- Confirm the requested model and effort are available. If the launch is rejected or the model is unavailable, explain that limitation; do not silently choose another model or create a separate user-owned chat as a substitute.
- Run no more than two Pro researchers at once for the task. An agent that is quiet or waiting still occupies a slot. Do not have them spawn additional researchers.
- Use two only for genuinely different questions or useful independent perspectives. Avoid duplicate investigations. A small edit or skill-maintenance task does not need a research agent.
Give one simple assignment
Write a short, direct prompt: what to investigate, the context needed to understand it, and what would help you next. Usually one paragraph is enough. Do not build a layered prompt with elaborate roles, mandatory report templates, or a prescribed conclusion. Give the researcher room to discover something you did not anticipate.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- yesterday First seen · 54 lines · 51 tokens per session scan A a6243029f809
pro-researcher is a skill published in the GitHub repository miuuyy/codex-chatgpt-web (14,335 stars, last pushed yesterday), licensed MIT. It adds 51 tokens to every session and 1,177 once invoked, about $0.0002 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 2026-10-15.
Other skills, from other repositories
bootstrap-realtime-eval
Bootstrap a new realtime eval folder inside this cookbook repo by choosing the right harness from examples/evals/realtimeevals, scaffolding prompt/tools/data files, generating a useful README, and validating it with smoke, full eval, and test runs. Use when a user wants to start a new crawl, walk, or run realtime eval…
expense-review-policy
Review invoices and contracts against accounts-payable policy before human approval.
auto-research
Research uncertain questions with an explicit, user-approved web search or ChatGPT consultation, then present options and wait for implementation approval.
powerpoint
Create designed, editable PowerPoint .pptx presentations with PptxGenJS. Use when the user asks to create, generate, update, or inspect a deck, slide deck, presentation, or .pptx file.
tmux
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
clawhub
Use the ClawHub CLI to search, install, update, and publish agent skills from clawhub.com. Use when you need to fetch new skills on the fly, sync installed skills to latest or a specific version, or publish new/updated skill folders with the npm-installed clawhub CLI.