REDDIT · REDDIT// 11d agoOPENSOURCE RELEASE

Claude Code-Inspired Deep Researcher Streamlines Literature Reviews

Deep Researcher is a Python, terminal-based academic research assistant that uses an agentic loop to search six scholarly databases, follow citation chains, refine queries, and generate literature reviews with BibTeX and structured output files. The project is explicitly modeled on Claude Code-style orchestration patterns, but applied to research workflows instead of coding, with support for local models as well as OpenAI- and Anthropic-compatible APIs.

// ANALYSIS

A strong niche open-source build: it takes a proven agent-loop design and applies it to academic research, where citation traversal and source coverage matter more than generic web summaries.

–Searches arXiv, Semantic Scholar, OpenAlex, CrossRef, PubMed, and CORE, which is materially better than web-only “deep research” tools for scholarly work.
–Outputs `report.md`, `references.bib`, `papers.json`, and `metadata.json`, so the result is usable in real writing workflows, not just a chat transcript.
–The architecture is intentionally lightweight: the repo claims about 1,500 lines and no LangChain, which should make it easier to understand and extend.
–The “leaked Claude Code framework” angle is more about borrowing agentic patterns than copying code, based on the README’s own framing.
–Main caveat: it is optimized for academic literature, not broad consumer web research, so the value depends on whether the user’s task fits that domain.

// TAGS

open-sourceacademic-researchliterature-reviewagentic-aiclaude-codepythonterminal-app

DISCOVERED

11d ago

2026-03-31

PUBLISHED

11d ago

2026-03-31

RELEVANCE

7/ 10

AUTHOR

plsendfast