Auto-Deep-Research: review, pricing, security and alternatives

What Auto-Deep-Research does

Auto-Deep-Research belongs to the deep research family: agents that search, read and cross-check dozens of sources to produce a sourced report. On our grid it ranks #18 of the 18 projects of this family listed in the directory.

You use it from: terminal (CLI). Usable models: Claude (Anthropic), GPT (OpenAI), DeepSeek, Mistral, Grok (xAI), local models (Ollama, llama.cpp, vLLM…), and any OpenAI-compatible model. It can run on your own machine or server.

Who is Auto-Deep-Research for?

It targets analysts, students and teams who must cross-check many sources to produce a reasoned summary. Typical tasks people give it:

• write a sourced overview of a topic
• compare products or competitors
• prepare a technical or regulatory watch

It is a good candidate if you are comfortable in a terminal, your data must stay on your machines, and you want the option of a local model, with no API fees.

Access and human control

According to its documentation, Auto-Deep-Research can:

• read and edit files
• drive a web browser
• call online services and APIs

We could not establish whether it asks for human approval before sensitive actions: test it first on an unimportant account or folder.

Auto-Deep-Research pricing

The code is public but the repository declares no licence: reuse is not formally allowed. The real cost comes from model calls: each loop step sends the whole context back to the model, and a task of several dozen steps can cost far more than a single question. The LLM price calculator estimates it for the chosen model.

Getting started with Auto-Deep-Research

1. Install it: the project ships as a Python package (install with pip or uv). The exact command is in the repository README.
2. Pick the model: an API key from a provider, or a local model if your machine has enough graphics memory (the VRAM calculator tells you).
3. Give it a first playground with nothing at stake: a test folder, a secondary account or a throwaway project. Only open your real accounts once you have seen how it behaves.
4. Give it a short, checkable task, review the result and its action log, then raise the difficulty step by step.

Security: risk level

An agent's report may cite misread or invented sources: check the links before reusing a figure.

Points found for Auto-Deep-Research:

• Outside content: it reads web pages or API responses, which may contain booby-trapped instructions (prompt injection).
• Maintenance: no commit for 12 months, flaws are no longer fixed.
• No licence: legally shaky for company use.

Three habits: least privilege (minimal rights, dedicated keys), human approval for irreversible actions, a sandbox for code execution. The AI agent security check reviews your setup in a few minutes.

Project activity and releases

Created on February 16, 2025, the HKUDS/Auto-Deep-Research repository has 1,752 stars, 246 forks and 33 open issues, with 18 commits in total. Last change 12 months ago.

Over the last 52 weeks: 0 commits, an average of 0 per week.

The project publishes no numbered releases on GitHub: you install it from the main branch or a package manager.

Auto-Deep-Research vs other deep research

GitHub stars of the best-rated open-source projects in the deep research family. Stars measure attention, not quality: a heavily promoted project collects them fast, a specialised tool will have few.

Auto-Deep-Research and other deep research at a glance

The 7 best-ranked agents in the same family, 7 of them open source. The score out of 100 only applies to open-source projects; for proprietary services we show the published entry price. A dash marks information the documentation does not state.

How we rated it

Score: 30/100, computed only from public, verifiable data, with no subjective testing or partnership:

• Adoption 15/40: 1,752 stars on a log scale (40 points from about 200,000).
• Activity 0/30: last commit 12 months ago, 0 releases in six months, 0 commits in a year.
• Framework 15/30: licence, default human approval, self-hosting, model choice and documentation site.

The score does not say whether the agent will succeed at your tasks: it measures how solid the project is. The full grid is on the method page.