explaingit

oh-rid/deep-research

23Audience · developerComplexity · 3/5Setup · moderate

TLDR

A Claude Code plugin that researches any question by querying Claude, Gemini, and GPT-5 in parallel, then cross-checks their answers and mechanically verifies every cited URL before including it in the report.

Mindmap

mindmap
  root((Deep Research))
    How It Works
      Three AI systems
      Parallel web search
      Claim comparison
    Verification
      URL fetch check
      CLI flag testing
      Hallucination flags
    Output
      Consensus claims
      Disagreements
      Source URLs
      Confidence ratings
    Safeguards
      Isolated file system
      Untrusted web data
      Rejected claims log
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Research a technical question and get a verified report showing where all three AIs agreed versus where they diverged

USE CASE 2

Catch AI hallucinations on command-line flags or API syntax by having the plugin verify claims against live tool output

USE CASE 3

Get a confidence-rated report with working source URLs for any topic where a single AI might confidently give wrong answers

Tech stack

Claude APIGemini APIGPT-5Claude Code

Getting it running

Difficulty · moderate Time to first run · 30min

Requires cloning into the Claude Code plugins folder and separately installing the Gemini and GPT-5 CLI cross-checkers for full three-way comparison.

In plain English

This is a plugin for Claude Code, a command-line AI coding tool, that runs research questions through three separate AI systems at once instead of just one. The idea is that any single AI can confidently state wrong things, especially when its training data overlaps with other AIs and they all learned the same incorrect information from the same sources. By running three different AI families in parallel (Claude, Google's Gemini, and OpenAI's GPT-5) and comparing what they say, the plugin has a better chance of catching errors. When you type /research followed by a question, the plugin spawns a researcher agent that queries all three AI systems using their own separate web search tools. Each system searches the web independently. After all three return answers, the researcher sorts the claims into categories: things all three agreed on, things only one said, and things where they disagreed. Before including any claim in the final report, the plugin checks whether a real URL supports it, fetching the page and verifying that the quoted passage actually appears there word for word. The plugin is designed to defend against several specific failure modes. Fabricated citations, where an AI confidently cites a paper or URL that does not exist or does not say what was claimed, are caught by the mechanical URL check. Claims about command-line tool flags or API syntax are verified by actually running the tool's help output rather than trusting the AI's memory. Content fetched from web pages is treated as untrusted data to prevent malicious instructions on a page from hijacking the research session. The entire researcher process runs in an isolated copy of the project so it cannot touch your actual files. The final report includes a summary, numbered findings with source URLs, a confidence rating per topic, claims that were rejected after verification, and a flag for claims where all three AIs agreed but no primary source could be found, which the plugin treats as a higher hallucination risk rather than a confirmation. Installation involves cloning the repository into a local Claude Code plugins folder and restarting the tool. The two optional cross-checker CLIs must be installed separately.

Copy-paste prompts

Prompt 1
Using the deep-research Claude Code plugin, research: What are the current best practices for React Server Components in Next.js 15? Show me consensus claims, disagreements between Claude, Gemini, and GPT-5, and verified source links.
Prompt 2
Run /research on 'what Python version should I target for a new production service in 2025' and give me the cross-AI consensus report with confidence ratings and rejected claims.
Prompt 3
I want to install the deep-research Claude Code plugin. Walk me through cloning the repo into my plugins folder and installing the optional Gemini and GPT-5 cross-checker CLIs.
Prompt 4
How does deep-research verify that a cited URL actually contains the quoted claim? Walk me through the mechanical URL-check step and what counts as a verification failure.
Open on GitHub → Explain another repo

← oh-rid on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.