ttguy0707

ttguy0707 / IRIS

Public

IRIS 是一个基于 Agentic Workflow（智能体工作流）的自动化深度调研与报告生成系统。它摒弃了传统的单向 RAG 问答模式，通过构建多节点状态机（State Machine），实现了从意图识别、路径规划、动态检索（混合/本地）、深度撰写到自我审查与局部修改的全自动闭环。

agent langgraph llm python rag
25
4
69% credibility
Found Feb 22, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

IRIS is a web app that generates detailed research reports by processing user-uploaded PDFs and optionally searching the web through an AI-powered workflow with planning, research, writing, and review steps.

How It Works

1
🔍 Discover IRIS

You stumble upon IRIS, a smart helper that turns your questions into detailed reports, right from your web browser.

2
🖥️ Open the App

Load the beautiful IRIS page and feel the excitement of an easy-to-use research buddy waiting for you.

3
📁 Add Your Documents

Drag and drop up to five PDF files into the cozy upload spot to build your personal knowledge base.

4
Pick Your Search Style
📄
Just My Docs

Stick to only the files you uploaded for private, focused insights.

🌐
Docs Plus Web

Mix your documents with fresh internet finds for richer, complete answers.

5
🚀 Ask Your Question

Type in your research topic like 'What are AI trends in 2026?' and hit the big button – magic starts happening live!

6
📊 Watch It Work

Follow the glowing steps on screen as it plans, searches, writes, and checks until perfect.

🎉 Read Your Report

Sit back and enjoy a beautifully typed, insightful report full of facts and analysis, ready to use.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 25 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is IRIS?

IRIS builds automated research reports from user queries using Python FastAPI backend and Vue frontend. Upload up to 5 PDFs for a local knowledge base, toggle doc-only or hybrid web search via Tavily, then hit "Initiate Research" for a streaming agentic workflow that plans paths, gathers data, drafts content, self-reviews, and iterates until polished. Ditch one-shot RAG chats for closed-loop depth—not minecraft iris github shaders or iris agentic ai clones like iris github csv/dataset.

Why is it gaining traction?

Real-time status flow, terminal logs, and typewriter report rendering make workflows tangible and fun to watch. Hybrid retrieval smartly blends docs/web, auto-switches on irrelevance, and QA loops fix drafts without manual tweaks. Devs grab it for quick agentic prototypes over raw LangGraph boilerplate.

Who should use this?

Analysts turning scattered PDFs/web into executive summaries. Indie researchers on trends like iris berben alter or irischer wolfshund breeds. AI tinkerers needing a runnable iris github amd-style agent without setup hell.

Verdict

Solid starter for agentic research at low cost, but 17 stars and 0.7% credibility signal prototype risks—sparse tests, API keys needed (Tavily/OpenAI). Run locally if you hack workflows; skip for production.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.