Claude Code for Academic Research

A complete Claude Code infrastructure for PhD researchers — skills, agents, hooks, and rules for academic workflows. Built for researchers who write papers in LaTeX, manage bibliographies, run experiments, and want AI assistance that understands academic conventions.

What's Included

Component	Count	Description
Skills	31	Slash commands for common tasks (/proofread, /latex-autofix, /literature, etc.)
Agents	6	Specialised reviewers (peer review, referee 2, proposal review, paper critic, domain review, fixer)
Hooks	8	Automated guardrails (destructive git protection, context monitoring, source file protection, etc.)
Rules	9	Always-on policies (plan before implementing, scope discipline, design before results, etc.)
Context library	—	Structured files that give Claude persistent memory about you, your projects, and your workflows
OpenAlex MCP	—	Scholarly search server for academic literature discovery

Quick Start

1. Clone

git clone https://github.com/flonat/claude-code-public.git
cd claude-code-public

2. Run setup

./scripts/setup.sh

This creates symlinks so Claude Code can find your skills, agents, hooks, and rules from any project directory.

3. Customise your context

Edit these files with your own details:

• .context/profile.md — Your name, institution, research areas, supervisors
• .context/current-focus.md — What you're working on right now
• .context/projects/_index.md — Your active research projects
• CLAUDE.md — Conventions and tool preferences

4. Configure settings

Edit .claude/settings.json to adjust:

• Allowed/denied commands
• Hook configuration
• Model preferences

5. Start using it

cd ~/your-research-project
claude

Claude will automatically load your context, skills, and rules. Try:

• /proofread — Academic proofreading for a LaTeX paper
• /latex-autofix — Compile LaTeX with automatic error fixing
• /validate-bib — Check citation keys against your .bib file
• /literature — Search for academic papers and manage bibliography
• /session-log — Create a timestamped progress log
• /code-review — 11-category quality review for R/Python scripts

Project Structure

claude-code-public/
├── CLAUDE.md                    # Main instruction file (customise this)
├── README.md                    # This file
├── MEMORY.md                    # Accumulated knowledge (auto-populated)
├── .claude/
│   ├── agents/                  # 6 specialised review agents
│   ├── rules/                   # 9 auto-loaded policy rules
│   └── settings.json            # Permissions, hooks, model config
├── skills/                      # 31 slash commands
│   ├── shared/                  # Shared utilities (palettes, scoring, rhetoric)
│   ├── proofread/               # Academic proofreading
│   ├── latex-autofix/           # LaTeX compilation + auto-fix
│   ├── literature/              # Literature search + synthesis
│   └── ...                      # See docs/skills.md for full list
├── hooks/                       # 8 automated guardrails
├── .context/                    # AI context library
│   ├── profile.md               # Your identity and background
│   ├── current-focus.md         # What you're working on NOW
│   ├── projects/                # Project metadata
│   ├── preferences/             # Workflow preferences
│   ├── workflows/               # Process guides (daily review, etc.)
│   └── resources/               # Reference data (journal rankings, etc.)
├── .mcp-server-openalex/        # OpenAlex scholarly search server
├── docs/                        # Component documentation
├── log/                         # Session logs (auto-created)
└── scripts/
    └── setup.sh                 # Initial setup script

Key Concepts

Context Library (.context/)

The context library gives Claude persistent memory about you and your work. Instead of re-explaining your research every session, Claude reads these files automatically.

• profile.md — Who you are (background, supervisors, tools)
• current-focus.md — What you're doing right now (updated each session)
• projects/_index.md — All your research projects at a glance
• workflows/ — How you want Claude to help with recurring tasks
• preferences/ — Your priority framework and naming conventions

Skills (skills/)

Skills are slash commands — type /proofread and Claude runs a structured academic proofreading protocol. Each skill has a SKILL.md file defining its behaviour, and optionally a references/ directory with supporting material.

See docs/skills.md for the full catalogue.

Agents (.claude/agents/)

Agents are specialised personas for complex review tasks. They use multi-agent architectures (spawning sub-agents for parallel work) and produce formal reports.

Agent	Use case
referee2-reviewer	Rigorous audit of your own work (code, methods, replication)
peer-reviewer	Review someone else's paper (spawns citation, novelty, methods sub-agents)
proposal-reviewer	Review research proposals and extended abstracts
paper-critic	Adversarial LaTeX paper audit with quality scoring
domain-reviewer	Substantive correctness (math, assumptions, code-theory alignment)
fixer	Apply fixes from a critic report

See docs/agents.md for detailed descriptions.

Hooks (hooks/)

Hooks run automatically at specific points in a Claude Code session to provide guardrails and context.

Hook	Trigger	What it does
block-destructive-git.sh	Before Bash	Catches dangerous git commands (force push, reset --hard, etc.)
context-monitor.py	After tool use	Tracks context usage and warns before auto-compact
promise-checker.sh	Session stop	Catches Claude saying "I'll remember" without actually writing
startup-context-loader.sh	Session start	Loads context files automatically
resume-context-loader.sh	Session resume	Restores focus on resume
precompact-autosave.py	Before compact	Saves state before context compression
postcompact-restore.py	After compact	Restores state after context compression
protect-source-files.sh	Before edit/write	Warns when editing files outside the project

See docs/hooks.md for full documentation.

Rules (.claude/rules/)

Rules are always-on policies that Claude follows in every session. They enforce good research practices.

See docs/rules.md for full documentation.

Customisation

Adding a new research project

1. Create a directory for your project
2. Add a CLAUDE.md in the project with project-specific instructions
3. Symlink paper/ to your Overleaf directory (if applicable)
4. Add the project to .context/projects/_index.md
5. Create a paper file in .context/projects/papers/

Adding your own skills

Create a new directory in skills/ with a SKILL.md file:

skills/my-custom-skill/
└── SKILL.md

The SKILL.md needs YAML frontmatter with name, description, and optionally allowed-tools. See any existing skill for the format.

Setting up OpenAlex MCP

The OpenAlex MCP server provides scholarly search capabilities:

1. Edit .mcp-server-openalex/server.py — update the email address
2. Add the server to your Claude Code MCP configuration
3. Use /literature to search for papers

Requirements

• Claude Code (CLI)
• LaTeX distribution (for paper compilation skills)
• uv (for Python package management)
• Git

Credits

This infrastructure draws on design patterns from Pedro Sant'Anna's Claude Code workflow (specialist agents, plan-first, quality gates, critic-fixer loops, [LEARN] tags).

License

MIT