skills.rest
mcncarl
mcncarlCOMMUNITY
Explore all skills by this author →
GitHub

yichen-chatgpt-web-research

Runs research tasks through the signed-in ChatGPT website and saves verified Markdown reports.

2.0k|274|Updated Feb 11, 2026
Search Tags:#web research#computer use#chatgpt#Browser Automation#Chrome extension#product research#Markdown report
One-click install
npx skills add https://github.com/mcncarl/yichen-skills --skill yichen-chatgpt-web-research
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: yichen-chatgpt-web-research
Source: https://github.com/mcncarl/yichen-skills/tree/main/yichen-chatgpt-web-research
Command: npx skills add https://github.com/mcncarl/yichen-skills --skill yichen-chatgpt-web-research

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It lets you run product, market, and competitor research through the official ChatGPT web page using your already signed-in account (including GPT-5.5 Pro), avoiding API costs and guaranteeing the answer comes from the real ChatGPT website rather than a substitute source.

Core Features & Use Cases

  • Real ChatGPT Web Control: Operates the actual chatgpt.com page in your logged-in Chrome profile via the Chrome extension route, with Computer Use as a visible-window fallback.
  • Verified Completion: Requires a unique completion marker in the response, polls long-running Pro generations, and extracts the answer via copy button, DOM, or accessibility tree.
  • Verified Markdown Reports: Saves raw output and a readable report with conversation URL, extraction method, and verification status.
  • Use Case: Ask it to research a competitor product through ChatGPT 官网 with your Pro account; it controls the real page, waits for the full answer, and saves a structured Chinese Markdown report to your workspace.

Quick Start

Use $yichen-chatgpt-web-research to research a product through the official ChatGPT web page and save a verified Markdown report.

Frequently Asked Questions about yichen-chatgpt-web-research

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run ChatGPT web research without using the OpenAI API?▼

This skill controls the real chatgpt.com page in your already signed-in Chrome profile, submits the research prompt, waits for completion, and extracts the answer from the page. It never falls back to the OpenAI API or generic web search.

How to use ChatGPT Pro or GPT-5.5 Pro for product research from Claude Code?▼

The skill locates your logged-in Pro Chrome profile, confirms visible Pro or GPT-5.5 labels on the page, submits a structured research prompt, and saves the verified Markdown report. If Pro status cannot be confirmed, it stops before sending.

What happens if Chrome extension control is unavailable?▼

It falls back to Computer Use on the real visible Chrome window, operating the same signed-in ChatGPT tab through screenshots and accessibility-tree extraction. If neither route can operate the intended page, it stops rather than substituting another source.

Does this skill inspect cookies or browser session data?▼

No. The skill explicitly forbids inspecting cookies, local storage, passwords, browser profiles, or session stores. It uses only visible page information and tab metadata to verify the account and model.

Why does the skill stop instead of answering when the ChatGPT page is unavailable?▼

Its hard rule requires answers to come from the real ChatGPT website in the intended account. If login, CAPTCHA, a free account, or missing Pro access blocks the task, it asks you to fix the browser state instead of producing a substitute answer.

Related Skills

Explore More Education & Research Skills→
krishna-cyberkrishna-cyber

browser-debug-setup

Configure a dedicated Chrome profile with remote debugging flags.

Community
Intermediate
ammiforuammiforu

save-article

Convert web articles into Obsidian-ready Markdown notes with YAML frontmatter.

Community
Intermediate
vasilievyakovvasilievyakov
17

triangulate

Cross-check claims against three or more independent sources and produce a structured verdict.

Community
Intermediate
GGEduGGEdu

cost-aware-llm-pipeline

Route LLM tasks to cost-appropriate models and track cumulative API spend.

Community
Advanced