page-monitoring
Claude Code skills for journalism, media, and academia - verification, FOIA, data journalism, academic writing, and more
npx skills add https://github.com/jamditis/claude-skills-journalism --skill page-monitoringInstale esta skill com a CLI e comece a usar o fluxo de trabalho SKILL.md em seu espaço de trabalho.
Claude skills for journalism, media, and academia
A curated collection of Claude Code skills designed for journalists, researchers, academics, media professionals, and communications practitioners.
Docs site: jamditis.github.io/claude-skills-journalism — interactive skill browser, setup guides, and full documentation.
Guides
Setup and workflow guides, separate from the skills themselves:
| Guide | Description |
|---|---|
| Persistent sessions | Keep Claude Code sessions alive through disconnects using tmux — setup, key bindings, activity notifications, and scheduler coexistence |
What are Claude skills?
Skills are modular instruction sets that extend Claude's capabilities for specialized tasks. Each skill contains domain-specific knowledge, workflows, templates, and best practices that Claude loads automatically when relevant to your work.
Installation
Prerequisite: You need Claude Code installed. Run claude --version in your terminal to check.
Plugins (recommended)
Plugins give you slash commands you can run directly inside Claude Code. Run these two commands in your terminal:
claude plugin marketplace add https://github.com/jamditis/claude-skills-journalism
claude plugin install pdf-playground@claude-skills-journalism
Then restart Claude Code (close and reopen). See the PDF Playground README for detailed setup instructions and troubleshooting.
Available plugins:
| Plugin | Description | Commands |
|---|---|---|
| autocontext | Cross-session knowledge persistence with skill evolution. Lessons accumulate per-skill, and /autocontext:evolve folds them back into skill files |
/autocontext:setup, /autocontext:init, /autocontext:review, /autocontext:status, /autocontext:evolve |
| dev-toolkit | Ten development-focused skills for journalists, researchers, and small newsroom dev teams: accessibility (WCAG 2.2), Electron app patterns, mobile/remote debugging, irreversible-decision discipline, Python data pipelines, test-first bug fixing, AI-assisted development workflows, ethical web scraping, no-build frontend patterns, and signs-of-taste guidance for web UI | n/a — skills only |
| journalism-core | Thirteen core journalism skills, including AP-style writing, AI-slop detoxing, source verification (deepfakes/C2PA), FOIA + NJ OPRA requests, fact-checking, interview prep + transcription, story pitches, editorial workflow, crisis communications, and newsletter publishing with current Gmail / Yahoo / Outlook bulk-sender requirements | n/a — skills only |
| pdf-design | PDF report and proposal design system with brand variables, budget tables, and reusable content blocks (stats strips, three-column, four-tile pillars, partner grids) | n/a — skill-only |
| pdf-playground | Create branded proposals, reports, one-pagers, newsletters, slides, and event materials with an interactive control panel for live design editing (colors, fonts, spacing, sections) and a guided wizard for proposals | /pdf-playground:proposal, /pdf-playground:report, /pdf-playground:onepager, /pdf-playground:newsletter, /pdf-playground:slides, /pdf-playground:event, /pdf-playground:preview |
| research-toolkit | Five skills for research, source preservation, and academic workflows: academic writing, legal paywall-bypass via Unpaywall and library databases, web archiving (Wayback / Archive.today / ArchiveBox), web page change monitoring, and AI-enriched digital archive construction | n/a — skills only |
| security-toolkit | Three defensive web application security skills covering OWASP Top 10 fundamentals: pre-deployment audit checklists (auth, input validation, secrets management), secure authentication implementation patterns (password hashing, session management, JWT, OAuth, passkeys), and API hardening (rate limiting, CORS, request throttling, defense-in-depth for Express, FastAPI, and serverless) | n/a — skills only |
| superjawn | Research-augmented fork of obra/superpowers. Default-on research phase fires before brainstorming, systematic-debugging, and writing-skills. v1.0.0 ships all 14 skills with no soft dependencies on the upstream superpowers plugin |
invoked indirectly via skills (e.g. superjawn:brainstorming, superjawn:systematic-debugging, superjawn:writing-plans) |
Skills (manual installation)
Most skills now ship inside plugins (see the table above). For skills that haven't been bundled into a plugin yet (e.g., free-apis-catalog, visual-explainer, project-memory), Claude Code discovers skills at ~/.claude/skills/<skill-name>/SKILL.md — one level deep. Clone this repo anywhere you like, then copy or symlink the bare skills you want:
git clone https://github.com/jamditis/claude-skills-journalism.git ~/projects/claude-skills-journalism
cd ~/projects/claude-skills-journalism
mkdir -p ~/.claude/skills
# Copy a bare top-level skill:
cp -r free-apis-catalog ~/.claude/skills/
# Or pull a single skill out of a plugin's skills/ directory:
cp -r journalism-core/skills/source-verification ~/.claude/skills/
cp -r research-toolkit/skills/web-archiving ~/.claude/skills/
cp -r dev-toolkit/skills/web-scraping ~/.claude/skills/
cp -r security-toolkit/skills/secure-auth ~/.claude/skills/
# Or symlink so git pull updates them in place (ln -sfn replaces an existing link):
ln -sfn "$PWD/free-apis-catalog" ~/.claude/skills/free-apis-catalog
Do not clone the repo directly into ~/.claude/skills/journalism-skills/ — that nests each SKILL.md too deep and Claude Code won't find them.
For Claude.ai users
Skills can be added via the Claude.ai interface under Settings > Skills.
Skills overview
Core journalism skills (in journalism-core plugin)
These thirteen skills ship together as the journalism-core plugin. Install via /plugin install journalism-core@claude-skills-journalism to get all of them at once.
| Skill | Description |
|---|---|
| ai-writing-detox | Eliminate AI-generated patterns that erode reader trust. Banned words, phrases, and structures with alternatives |
| crisis-communications | Breaking news protocol, rapid verification, crisis response, misinformation countering |
| data-journalism | Dataset analysis, chart and map creation, statistical reasoning, data-driven story structure |
| editorial-workflow | Story assignment tracking, deadline management, editorial calendars, handoff protocols |
| fact-check-workflow | Claim extraction, evidence gathering, rating scales, correction protocols |
| foia-requests | Public records request drafting (federal FOIA + NJ OPRA), tracking, appeals, current statutory citations |
| interview-prep | Pre-interview research, question frameworks, recording consent, attribution guidelines |
| interview-transcription | Whisper / WhisperX transcription pipelines, quote management, speaker diarization |
| newsletter-publishing | Email newsletter creation, subscriber management, deliverability, 2024–2026 Gmail / Yahoo / Outlook bulk-sender compliance |
| newsroom-style | AP Style enforcement, attribution rules, headline formatting, number conventions |
| social-media-intelligence | Narrative tracking, coordinated-campaign analysis, account authenticity checks, OSINT for digital investigations |
| source-verification | SIFT method, image and video verification, deepfake detection (2026), C2PA Content Credentials, verification trails |
| story-pitch | Pitch templates for daily news, features, investigations, op-eds, and freelance queries |
Research and academic skills (in research-toolkit plugin)
These five skills ship together as the research-toolkit plugin. Install via /plugin install research-toolkit@claude-skills-journalism to get all of them at once.
| Skill | Description |
|---|---|
| academic-writing | Research design, literature reviews, IMRaD structure, peer review responses, grant proposals |
| content-access | Unpaywall, CORE, Semantic Scholar APIs, library databases, ethical access patterns |
| digital-archive | Building archives with AI enrichment, entity extraction, knowledge graphs |
| page-monitoring | Change detection, RSS generation, webhook alerts, automatic archiving on changes |
| web-archiving | Wayback Machine, Archive.today, legal evidence preservation, multi-archive redundancy |
Design and production
| Skill | Description |
|---|---|
| pdf-design | Professional PDF reports and proposals with brand system, budget tables, and multi-page layouts. For the full interactive experience, use pdf-playground instead |
| visual-explainer | Turn complex data into styled HTML pages — architecture diagrams, data tables, flowcharts, timelines, source maps, and dashboards with dark/light theme support |
Project documentation
| Skill | Description |
|---|---|
| project-memory | Generate CLAUDE.md files that capture project-specific knowledge. Includes templates for editorial tools, events, publications, research, pipelines, and archives |
| project-retrospective | Generate LESSONS.md files that document what worked and what didn't. Templates for investigations, events, publications, and tools |
| template-selector | Decision tree for choosing the right project documentation template |
Development skills (in dev-toolkit plugin)
These ten skills ship together as the dev-toolkit plugin. Install via /plugin install dev-toolkit@claude-skills-journalism to get all of them at once, or copy individual skills from dev-toolkit/skills/<name>/.
| Skill | Description |
|---|---|
| accessibility-compliance | WCAG 2.2 baseline, alt text, focus management, motion preferences, accessible charts |
| electron-dev | Electron security model (contextIsolation, sandbox), IPC patterns, packaging |
| mobile-debugging | Eruda, vConsole, Chrome DevTools on Android, Safari Web Inspector for iOS, console capture |
| one-way-door | Flag irreversible architectural decisions (data models, infra, auth, APIs) before committing |
| python-pipeline | Data pipelines (pandas, polars, DuckDB, asyncio) with modular architecture |
| test-first-bugs | Test-driven bug fixing: write failing test first, then fix with subagents |
| vibe-coding | AI-assisted development workflow (Claude Code, Cursor, Aider, Continue) |
| web-scraping | Ethical scraping patterns (Playwright, robots.txt, anti-bot defense, terms-of-service) |
| web-ui-best-practices | Container queries, :has(), view transitions, scroll-driven animations, signs of taste in web UI |
| zero-build-frontend | ESM import maps, htmx, Alpine.js, no-build deployment |
Security skills (in security-toolkit plugin)
These three skills ship together as the security-toolkit plugin. Install via /plugin install security-toolkit@claude-skills-journalism to get all of them at once.
| Skill | Description |
|---|---|
| api-hardening | Rate limiting, input validation, CORS, security headers, API key management, defense-in-depth for Express, FastAPI, and serverless |
| secure-auth | Production-ready authentication patterns: password hashing, session management, JWT, OAuth, passkeys / WebAuthn, MFA |
| security-checklist | Pre-deployment security audit covering OWASP Top 10 fundamentals: authentication, input validation, secrets, database security, compliance basics |
Reference
| Skill | Description |
|---|---|
| free-apis-catalog | 1000+ free public APIs by category (finance, news, weather, sports, ML, entertainment) with evaluation criteria and project ideas |
Hooks
Hooks are automated checks that run at specific points in your workflow. All hooks are non-blocking warnings—they provide guidance but don't prevent actions.
Writing quality hooks
| Hook | Event | Description |
|---|---|---|
| ap-style-check | PostToolUse | Flag common AP Style violations |
| ai-slop-detector | PostToolUse | Warn about AI-generated patterns |
| accessibility-check | PostToolUse | Check alt text, heading structure, link text |
Verification hooks
| Hook | Event | Description |
|---|---|---|
| source-attribution-check | PostToolUse | Flag unattributed quotes and claims |
| verification-reminder | PostToolUse | Prompt to verify facts before including |
| data-methodology-check | PostToolUse | Ensure data stories include methodology |
Editorial workflow hooks
| Hook | Event | Description |
|---|---|---|
| source-diversity-check | PostToolUse | Note when sources may lack diversity |
| legal-review-flag | PostToolUse | Flag potentially defamatory content |
| pre-publish-checklist | Stop | Reminder checklist before completing tasks |
| deadline-tracker | SessionStart | Surface upcoming deadlines |
Preservation hooks
| Hook | Event | Description |
|---|---|---|
| archive-reminder | PostToolUse | Remind to archive URLs when citing sources |
Development hooks
| Hook | Event | Description |
|---|---|---|
| one-way-door-check | PreToolUse | Block creation of files representing irreversible architectural decisions |
| bug-report-detector | UserPromptSubmit | Detect bug reports and remind to follow test-first workflow |
| enforce-test-first | PreToolUse | Block source code edits until a test file has been written |
Skill structure
Each skill follows the Claude Agent Skills standard:
skill-name/
├── SKILL.md # Main instructions (required)
├── examples/ # Example inputs/outputs (optional)
├── templates/ # Reusable templates (optional)
└── scripts/ # Helper scripts (optional)
SKILL.md format
---
name: skill-name
description: When to use this skill and what it does
---
# Skill title
Instructions and knowledge for Claude to use.
Usage examples
Source verification
When you ask Claude to verify a claim or check a source:
"Can you help me verify this viral tweet claiming [X]?"
"What steps should I take to verify this document?"
"Help me check if this image is authentic"
Interview preparation
When preparing for interviews:
"Help me prepare questions for interviewing the mayor about the budget"
"What background research should I do before this interview?"
"Create a question framework for an investigative interview"
Story pitching
When developing story ideas:
"Help me pitch this feature story idea"
"Draft a query letter for this freelance investigation"
"What's the angle for this daily news story?"
Fact-checking
When verifying claims:
"Walk me through fact-checking this politician's statement"
"Help me document the evidence for this claim"
"What rating should this fact-check receive?"
Project documentation
When setting up or closing projects:
"Generate a CLAUDE.md for this investigation project"
"Write a retrospective for the conference website we just finished"
"Which template should I use for this newsletter project?"
Social media intelligence
When investigating online narratives:
"Help me analyze this account for authenticity"
"Track how this claim is spreading across platforms"
"Check for signs of coordinated behavior"
Web archiving and monitoring
When preserving or tracking content:
"Archive this page to multiple services for redundancy"
"Set up monitoring for changes to this government page"
"Find academic papers related to this topic"
Contributing
Contributions welcome! See CONTRIBUTING.md for the full guide — skill structure, testing, and style guidelines.
Quick version: fork, create your skill with a SKILL.md frontmatter file, test with Claude Code, submit a PR.
Target audiences
- Investigative journalists - source verification, FOIA, data analysis
- Newsroom developers - scraping, data pipelines, visualization tools
- Academic researchers - writing, methodology, peer review
- Graduate students - thesis writing, literature reviews, proposals
- Communications professionals - content strategy, research methods
- Fact-checkers - verification workflows, evidence documentation
- Digital archivists - preservation, metadata, knowledge graphs
- Editors - workflow management, style enforcement, quality control
Related resources
- Anthropic official skills
- Claude Code documentation
- Agent Skills Standard
- NICAR (Investigative Reporters & Editors)
- First Draft News
- Verification Handbook
- AP Stylebook
- Bellingcat OSINT guides
License
MIT License - See LICENSE for details.
Author
Joe Amditis (@jamditis)
Built with Claude Code