Analysis updated 2026-05-18
Draft and revise a paper targeting a CCF-A conference or SCI journal with specialist AI agents.
Get an automated review of your manuscript's logical and structural problems before submission.
Search Semantic Scholar for supporting literature relevant to your paper's problem statement.
Convert reference PDFs into clean Markdown for the AI agents to read.
| reyuyu/ccfa-gent | audemodo/audemodo-responsive-keepalive | cahn91/gpt-repo-mcp | |
|---|---|---|---|
| Stars | 28 | 28 | 28 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | moderate | easy | moderate |
| Complexity | 3/5 | 2/5 | 3/5 |
| Audience | researcher | developer | vibe coder |
Figures from each repo's GitHub metadata at analysis time.
Requires Python 3.10+, Node.js 20, and a DeepSeek API key, Chrome or Edge recommended.
CCFA Paper Agent is a local AI writing workspace designed to help researchers write, check, and revise academic papers, particularly those targeting CCF-A conferences and SCI journals. It runs entirely on your own machine using a React web interface backed by a Python API server. The system organizes each paper as a project that holds your draft files, reference papers, figures, and notes about the scientific problem you are addressing. Rather than a single chatbot, it routes your requests to specialized agents depending on what you need: one agent handles writing and revision, another reviews your manuscript for logical and structural problems, a third searches academic literature on Semantic Scholar to find supporting papers, and a coordinator decides which specialist to call. A notable design choice is that no file on your machine is edited without your approval. When any agent proposes a change to your manuscript, it generates a diff that you can inspect in the browser before accepting or rejecting it. This means you can run the AI freely without worrying about it silently overwriting your work. Other features include PDF parsing via a tool called MinerU that converts reference papers into clean Markdown, a phrase memory that accumulates useful academic sentence patterns across sessions, and streaming output so you can watch the agents' reasoning steps in real time. Setup requires Python 3.10 or newer, Node.js 20, and a DeepSeek API key to power the agents. A Semantic Scholar API key is optional but improves rate limits for literature search. Running the provided startup script installs dependencies and opens the workspace in your browser at a local address. Chrome or Edge are recommended because the app uses browser file-system features that Safari does not fully support.
A local AI writing workspace with specialist agents that help researchers draft, review, and revise academic papers.
Mainly TypeScript. The stack also includes React, Python, DeepSeek.
Setup difficulty is rated moderate, with roughly 1h+ to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Verify against the repo before relying on details.