Skill Detail

Run local deep research workflows with Local Deep Research

Use Local Deep Research to run private agentic research passes across web, academic, and local document sources with citations and an encrypted knowledge base.

Research & ScrapingMulti-Framework
Research & Scraping Multi-Framework Security Reviewed
⭐ 7.9k GitHub stars
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill run-local-deep-research-workflows-with-local-deep-research Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Local Deep Research, Docker or Python runtime, Ollama or OpenAI-compatible LLM endpoint, search backend such as SearXNG, optional private document library
Install & setup
Install Local Deep Research with the official Docker, Docker Compose, or pip instructions, configure Ollama or another OpenAI-compatible LLM endpoint plus search services, then open the local web UI and run a research mode such as the LangGraph agent strategy.
Author
LearningCircuit
Publisher
Organization
Last updated
May 20, 2026
Quick brief

Use Local Deep Research when an operator needs a repeatable private research workflow that can run against local or cloud LLMs, multiple search engines, academic sources, and private documents. The workflow is to deploy Local Deep Research with Docker, Docker Compose, or pip, connect an LLM endpoint and search backend such as SearXNG, select a research mode such as the LangGraph agent strategy, and review the cited report plus saved research history.

How it works

What this skill actually does

Invoke this instead of using a hosted deep-research product when the task needs local control, encrypted per-user storage, private document search, reproducible source collection, or a local Ollama/llama.cpp/LM Studio-style model path. The scope boundary is agentic research execution and knowledge-base search; it is not a generic web search app, citation manager, or broad local-AI platform card.

Inputs are a research question, configured LLM provider or local model endpoint, search engines, optional private documents or indexed library sources, and runtime settings. Outputs are cited research reports, downloaded and indexed sources, searchable research history, and evidence that can be reviewed or rerun locally.