Skip to content

Productivity Blueprint

Safety checked

Run Academic Research Workflows With Claude Code

5flonatMay 2026

50 skills, 15 agents, 9 hooks, and 18 rules for researchers. Covers LaTeX authoring, bibliography management, code review, experimental design, and peer review.

Paper Writing & LaTeXBibliography & LiteratureCode Quality & ReproducibilityResearch Design & AnalysisReview & CritiqueProject Management & ContextSpecialized Audit & Verification

Install Prompt

Copy this prompt and paste it into your AI agent to install.

What problem does this solve?

Runs academic workflows end to end without switching between tools or repeating context. Skills handle LaTeX compilation, bibliography validation, code review, experimental design, and paper critique, so researchers work inside Claude Code instead of jumping between editors, terminals, and references.

How does it work?

Once installed, you access 50 slash commands from the terminal, VS Code, web, or desktop. You can start anywhere: proofread a paper, search literature, review code, plan an experiment, or validate citations. Skills read your context library (profile, project metadata, current focus) automatically, so each session builds on the last. Output: completed papers, validated references, reviewed designs, and audited code.

Included Skill Groups

What's the biggest win?

Research state and context stay in Claude Code instead of spread across files and tools, so every session knows what you did last time and what you're working on now.

What's required to run this?

Hooks run in Python 3.11+ or Bash and are configured via ~/.claude/settings.json. Skills use uv for Python dependency isolation instead of bare pip. The context library lives in .context/ and is symlinked into ~/.claude/ during setup, so all Claude Code interfaces (CLI, VS Code, web, desktop) share the same skills, agents, and rules.

What should I watch out for?

Requires Python 3.11+, TeX Live or MiKTeX for LaTeX skills, and API keys for optional MCP servers (Notion, bibliography providers). Some skills are read-only by design (proofread, review agents). The fixer agent applies changes only when explicitly invoked. Session continuity relies on you updating current-focus.md at the end of each session.

Key Features

50 Reusable Skills

Slash commands for proofread, LaTeX compile, literature search, bibliography validate, code review, experimental design, and more.

15 Specialized Agents

Read-only auditors for peer review, code-paper consistency, reproducibility, and manuscript critique that report findings without modifying source.

Context Library

Persistent markdown files (profile, projects, workflows, current focus) that give Claude access to your research state across all sessions.

9 Automated Hooks

Guardrails that run before/after key moments. Context monitoring, state preservation, git safety, file protection, and recovery.

18 Always-On Rules

Auto-loaded policies for research integrity: plan before implementing, audit before fixing, design before examining results, verify citations.

Tools in this Blueprint

Claude logo
4.4(68 reviews)
Claude Code (Terminal Mode)
Claude Code (terminal, VS Code, JetBrains, web, desktop)
Overleaf (LaTeX editor integration)

About This Blueprint

License
MIT
Industry
Technology
Skills
15 workflows, 27 sub-skills, 50 standalone
Sales

ICP-Filtered Outreach Lists from G2 Buyer Intent

Stop wasting rep time on weak intent signals and mismatched accounts. The workflow ranks ICP-fit buyers by readiness, finds the right contact, and supplies tailored outreach angles for the strongest opportunities.

ID

Ishani D.

Account Manager

Customer Experience

Answer pre-purchase questions to convert hesitant shoppers with AI

Shoppers get accurate answers about fit, compatibility, ingredients, and delivery even when your team is offline, so fewer leave before checkout. Conversation insights also reveal which product-page details could prevent the next unanswered question.

UB

Urska B.

GTM Lead

Customer Experience

Find the root cause behind repeat support tickets with AI

Catch a few unusual tickets before a product or fulfillment issue turns into hundreds of refunds, chargebacks, and bad reviews. Cluster transcript evidence by SKU, order date, location, and carrier so operations can trace the cause and support can reach affected customers early.

UB

Urska B.

GTM Lead