Summary

This repository provides a curated collection of "Claude Code skills" designed to enhance AI assistance for journalism, media, and academia. It offers specialized workflows, knowledge bases, and templates to improve tasks like source verification, data analysis, research, and content creation for professionals and researchers.

How It Works

Skills are modular instruction sets that extend Claude Code's capabilities. They are packaged into plugins (e.g., journalism-core, research-toolkit, dev-toolkit, security-toolkit) that Claude Code loads automatically. This approach allows for domain-specific enhancements, providing journalists, researchers, and developers with tailored AI support for complex tasks.

Quick Start & Requirements

• Prerequisite: Claude Code installed (claude --version).
• Installation: Recommended via Claude Code slash commands: /plugin marketplace add jamditis/claude-skills-journalism and /plugin install pdf-playground@claude-skills-journalism, followed by a restart. Manual installation involves cloning the repo and copying/symlinking skill folders to ~/.claude/skills/.
• Documentation: Interactive skill browser, setup guides, and full documentation available at skills.amditis.tech.

Highlighted Details

• Journalism & Research: Comprehensive skills for AP-style writing, AI-slop detoxing, source verification (deepfakes, C2PA), FOIA requests, fact-checking, interview prep, data journalism, academic writing, accessing paywalled content, and web archiving.
• Development & Security: Includes skills for WCAG 2.2 accessibility, Python data pipelines, ethical web scraping, modern frontend patterns, OWASP Top 10 security fundamentals, secure authentication, API hardening, and supply-chain hardening with install-time cooldowns.
• Productivity Tools: Features like pdf-playground for generating branded documents, autocontext for cross-session knowledge persistence, and numerous automated "hooks" for quality control (e.g., one-way-door-check, enforce-test-first, ap-style-check).

Maintenance & Community

Authored by Joe Amditis (@jamditis). The project appears actively maintained through its extensive catalog of skills and plugins. No specific community forums or sponsorship details are listed in the README.

Licensing & Compatibility

Released under the MIT License, which permits broad use, modification, and distribution, including for commercial purposes and integration into closed-source projects.

Limitations & Caveats

Strictly dependent on the Claude Code environment. Some advanced features, like deepfake detection, are noted with future-year markers (e.g., "2026"), indicating potential for ongoing development or reliance on future capabilities. The superjawn plugin is at v1.0.0, suggesting potential for API changes in future versions.