AlveeeRahman

Research Hound — Productivity skill for Claude Code

Productivity community

A scientific research workflow for Claude Code: five stages and 21 runnable scripts for citation verification, claim audits, manuscript linting, and rubric scoring.

How to install Research Hound

This entry records only its repository, not the path inside it, so there is no exact command to give. Open AlveeeRahman/research-hound and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Research Hound does

A scientific research workflow for Claude Code: five stages and 21 runnable scripts for citation verification, claim audits, manuscript linting, and rubric scoring.

Alternatives in Productivity

Embrace — Full Double Diamond workflow - Research → Define → Develop → Deliver 2.8k ★
Land A Job — Workflow recipe — go after a specific role end to end by chaining 4 skills: decode the JD, research the compan 1.3k ★
MCP Webresearch — MCP web research server (give Claude real-time info from the web) 298 ★

README

Research Hound: a scientific research workflow for Claude Code

[![CI](https://github.com/AlveeeRahman/research-hound/actions/workflows/ci.yml/badge.svg)](https://github.com/AlveeeRahman/research-hound/actions/workflows/ci.yml) [![SkillSpector Safety](https://github.com/AlveeeRahman/research-hound/actions/workflows/skillspector.yml/badge.svg)](https://github.com/AlveeeRahman/research-hound/actions/workflows/skillspector.yml) [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://github.com/AlveeeRahman/research-hound/blob/main/LICENSE) [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://www.python.org/downloads/)

**Documentation**: [alveeerahman.github.io/research-hound](https://alveeerahman.github.io/research-hound/) · part of a three-skill suite with [Skill Vision](https://alveeerahman.github.io/skill-vision/) and [Agent Oracle](https://alveeerahman.github.io/agent-oracle/).

**research-hound** is an [Agent Skill](https://code.claude.com/docs/en/skills) for Claude Code that turns Claude into a methodical research partner: a five-stage scientific workflow (critical thinking, literature review, brainstorming, writing, evaluation) backed by **21 runnable scripts** that verify what a chat model would otherwise just assert. It hunts down the weak citation, the untraceable claim, and the statistical pitfall before a reviewer does.

Automated safety score (SkillSpector)

This repository now runs [NVIDIA SkillSpector](https://github.com/NVIDIA/skillspector) on every push and pull request:

skillspector scan . --no-llm generates JSON + Markdown reports as workflow artifacts.
The workflow summary publishes the report's risk_score, severity, recommendation and the derived safe_to_install as proof of the current safety posture.
The check fails when safe_to_install is false, which uses SkillSpector's own rule: a risk score above 50, i.e. a DO_NOT_INSTALL recommendation. A CAUTION verdict pa