#125125th of 389 in Skills, plugins and promptsFor science, health and research, research and papers
hyperresearch
Deep research pipeline for Claude Code and Codex with a persistent source vault
- Stars
- 4.6k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- Python
39 on 21 Apr 20264,573 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Hyperresearch is a Python CLI that installs a 16-step research pipeline as skills and subagents for Claude Code and OpenAI Codex. It decomposes a prompt, fetches sources in parallel, drafts, runs four critics and a cite-checker, and applies surgical patches to produce a cited report. Fetched sources are stored in a markdown-plus-SQLite vault that later sessions search before fetching.
Who it is for: Engineers using Claude Code or Codex who want cited, auditable research reports
Strengths
- Cite-checker verifies sampled citation-sentence bindings; hallucinated quotes block the ship gate
- One scholarly search covers OpenAlex, Crossref, CORE, DOAB, ClinicalTrials.gov, SEC EDGAR and FRED
- Persistent vault and per-run manifest let later sessions reuse sources and resume crashed runs
- Tiers and gears scale runs from a ~30-minute light path to premier with 100-130 sources
Weaknesses
- Full runs take about 1.5-2.5 hours; premier runs take 3-5 hours
- Codex has no browser lane, and its tool locks are instructions, not enforcement
- Dissertation tier is experimental; its length and citation targets are not yet wired in
- Runs on top of Claude Code or Codex sessions, so model usage costs fall on the user
What it needs
- no GPU
- Needs Claude Code or OpenAI Codex CLI, Python 3.11-3.14, crawl4ai, SQLite
- Models: Claude (Sonnet, Opus, Haiku), OpenAI Codex models
Quick answers
Is hyperresearch open source?
Yes. Its code is on GitHub under the MIT license.
Also in Skills, plugins and prompts
See all 389| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | cavemanShortens coding-agent replies and compresses tool output before the model reads it | 95 out of 100 |