skills.rest
Ancienttwo
AncienttwoCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

repo-harness-chatgpt-browser

Consult logged-in ChatGPT Web sessions via the repo-harness browser engine.

407|29|Updated Mar 19, 2026
Search Tags:#planning#review#ChatGPT#repo-local workflow#browser consultation
One-click install
npx skills add https://github.com/Ancienttwo/repo-harness --skill repo-harness-chatgpt-browser
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: repo-harness-chatgpt-browser
Source: https://github.com/Ancienttwo/repo-harness/tree/main/.agents/skills/repo-harness-chatgpt-browser
Command: npx skills add https://github.com/Ancienttwo/repo-harness --skill repo-harness-chatgpt-browser

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill facilitates the consultation of a user's logged-in ChatGPT Web session via the repo-harness browser engine, allowing for planning, architecture review, PRD critique, and goal generation without relying on chat memory.

Core Features & Use Cases

  • Session Consultation: Accesses a user's ChatGPT Web session for planning and review tasks.
  • Dry-Run Capabilities: Executes planned actions without performing any real operations for validation.
  • Oracle Provider Consultation: Utilizes a provider for model selection and diagnostics.

Quick Start

To begin, use the 'repo-harness chatgpt browser-consult' command with your specific prompt and file, or run diagnostics using the 'repo-harness chatgpt browser-doctor' command.

Frequently Asked Questions about repo-harness-chatgpt-browser

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

FAQPage Schema
How do I use my logged-in ChatGPT Web session for PRD critique and planning?▼

You can use a browser engine to consult your logged-in ChatGPT Web session for PRD critique and planning by querying the session via an API interface. This requires authentication through a specific browser profile.

Can I run a dry-run validation for architecture review actions without executing real operations?▼

Yes, you can execute planned architecture review actions without performing real operations using dry-run capabilities. This validates your planned actions safely before any actual execution occurs.

Do I need a specific browser profile to consult ChatGPT Web sessions for goal generation?▼

Yes, consulting ChatGPT Web sessions for goal generation requires authentication via a browser profile. The browser engine operates by querying the authenticated session through a provided API interface.

How do I diagnose issues when consulting a logged-in ChatGPT Web session?▼

You can diagnose issues by running diagnostics using the browser-doctor command. This utilizes an Oracle provider for model selection and diagnostics to troubleshoot your ChatGPT Web session consultation.

What is the best way to generate goals from a ChatGPT Web session without relying on chat memory?▼

The best way to generate goals without relying on chat memory is using a browser engine to consult your logged-in ChatGPT Web session. It accesses the active session directly for planning and goal generation tasks.

Are there limitations when using a browser engine to access ChatGPT Web for review tasks?▼

A key limitation is that accessing ChatGPT Web for review tasks depends entirely on an active browser profile for authentication. You must maintain this session validity to ensure successful consultation.

Related Skills

Explore More Product & Management Skills→
esopianesopian
2

pair-agent

Generate setup keys to connect remote AI agents to browser sessions.

Community
Basic
incidentfoxincidentfox
654

remediation

Execute Kubernetes pod restarts, deployment scaling, and rollbacks with dry-run checks.

Official
Advanced
menoncellomenoncello
3

orchestrator-agent

Orchestrate subagents, commands, MCP servers, and skills with parallel execution.

Community
Intermediate
grandameniumgrandamenium
88

opencli

Convert logged-in Chrome sessions into repeatable CLI commands for structured data extraction.

Community
Intermediate