Projects - Deep Research Pipeline
Overview
Proposed Architecture
-
Local Deep Researcher
-
SearXNG
-
n8n
Components
Local Deep Researcher
SearXNG
-
searxng/searxng: SearXNG is a free internet metasearch engine which aggregates results from various search services and databases. Users are neither tracked nor profiled. - https://github.com/searxng/searxng
-
Why use a private instance? — SearXNG Documentation (2025.5.10+e982b9f73) - https://docs.searxng.org/own-instance.html
Logs
Setup
$ cd ~/public_git/ $ git clone https://github.com/langchain-ai/local-deep-researcher.git $ cd local-deep-researcher/ $ cp .env.example .env $ uvx --refresh --from "langgraph-cli[inmem]" --with-editable . --python 3.11 langgraph dev (When debugging in the home network:) $ uvx --refresh --from "langgraph-cli[inmem]" --with-editable . --python 3.11 langgraph dev --host 0.0.0.0
We have seen
(similar issue: https://community.openai.com/t/error-making-localhost-plugin-http-call-typeerror-failed-to-fetch/225524/2
Setup (SearXNG)
-
Docker Container — SearXNG Documentation (2025.5.10+e982b9f73) - https://docs.searxng.org/admin/installation-docker.html#installation-docker
$ mkdir -p ~/tmp/searxng/my-instance $ cd ~/tmp/searxng/my-instance (with modifying some) $ docker run --rm \ -d -p ${PORT}:8080 \ -v "${PWD}/searxng:/etc/searxng" \ -e "BASE_URL=http://localhost:$PORT/" \ -e "INSTANCE_NAME=my-instance" \ searxng/searxng
Referring SearXNGをDockerに載せてSearch API経由で検索する - https://zenn.dev/jun_philos/articles/28f137edc98cf3
(edit ls ~/tmp/searxng/my-instance/searxng/settings.yml) $ fgrep -n3 formats: ~/tmp/searxng/my-instance/searxng/settings.yml 74- recaptcha_SearxEngineCaptcha: 604800 75- 76- # remove format to deny access, use lower case. 77: # formats: [html, csv, json, rss] 78: formats: 79- - html 80- - json 81-
for other configurations, searxng/searxng-docker: The docker-compose files for setting up a SearXNG instance with docker. - https://github.com/searxng/searxng-docker or other resources may help.
Setup(Integrate with n8n)
-
LangChain learning resources | n8n Docs - https://docs.n8n.io/advanced-ai/langchain/langchain-learning-resources/
n8n-Based Chain
-
Experimenting with a Self-Hosted Deep Research Agent (n8n + SearXNG + Gemini 2.0) - Built with n8n - n8n Community - https://community.n8n.io/t/experimenting-with-a-self-hosted-deep-research-agent-n8n-searxng-gemini-2-0/89573
-
Host Your Own AI Deep Research Agent with n8n, Apify and OpenAI o3 | n8n workflow template - https://n8n.io/workflows/2878-host-your-own-ai-deep-research-agent-with-n8n-apify-and-openai-o3/?_gl=1*1pbwspp*_gcl_au*MjI2MDQ2NTY0LjE3NDYzNjUxNjc.*_ga*MTk5NTc1ODc0LjE3NDY4NzkyMjM.*_ga_0SC4FF2FH9*czE3NDY4NzkyMjMkbzEkZzEkdDE3NDY4ODIwODckajYwJGwwJGgw
-
References
-
langchain-ai/local-deep-researcher: Fully local web research and report writing assistant - https://github.com/langchain-ai/local-deep-researcher
-
langchain-ai/open_deep_research - https://github.com/langchain-ai/open_deep_research
-
OSS 版 DeepResearch を社内で試す ― local‑deep‑research|せきとば 〜ChatGPTと暮らす日々〜 - https://note.com/life_chatgpt/n/n3e873a2cc00e
-
SearXNG Tool node documentation | n8n Docs - https://docs.n8n.io/integrations/builtin/cluster-nodes/sub-nodes/n8n-nodes-langchain.toolsearxng/#running-a-searxng-instance
-
SearXNGをDockerに載せてSearch API経由で検索する - https://zenn.dev/jun_philos/articles/28f137edc98cf3
-
Setup your own safe search engine using SearXNG - DEV Community - https://dev.to/armiedema/setup-your-own-safe-search-engine-using-searxng-3428
https://github.com/langchain-ai/local-deep-researcher based
https://docs.searxng.org/own-instance.html
https://riq0h.jp/2023/08/20/091355/
-
Gemini 2.0 の Search as a tool を試す|npaka - https://note.com/npaka/n/n2cd8f9199a74
-
Xユーザーのうえぞう@うな技研代表さん: 「GeminiのGoogle Searchツールアホほど速いので、これをツール化してLLM問わず利用できるようにした」 / X - https://x.com/uezochan/status/1920421361081839743
https://docs.n8n.io/integrations/builtin/cluster-nodes/sub-nodes/n8n-nodes-langchain.toolsearxng/#running-a-searxng-instance
https://note.com/life_chatgpt/n/n3e873a2cc00e
r
ref