critical-article-writer

Generate draft articles, outlines, and editorial content matching a distinctive analytical, skeptical voice with sharp critical commentary, conversational to...

Security Scan

The skill's content matches its stated purpose (drafting critical articles), but the README mentions an external API and an API key (SkillBoss) that are not declared in the registry metadata — an inconsistency that could hide unexpected network usage or credential requirements.

Purpose & Capability

The name, description, and the instruction files are coherent: they provide templates, voice guidance, and prompt examples for generating critical articles and outlines. However, the README contains an unrelated integration example ('Installation via SkillBoss API Hub' with SKILLBOSS_API_KEY and https://api.heybossai.com/v1/pilot) that is not reflected in the skill's declared requirements. This is an unexplained mismatch between what the skill claims and additional suggested plumbing.

Instruction Scope

The SKILL.md and the instructions folder stay on-topic (prompts, templates, tone checklists) and do not instruct the agent to read local files, system paths, or arbitrary environment variables. They ask for evidence/sources to be provided in prompts, but do not themselves tell the agent to fetch external data or exfiltrate content.

Install Mechanism

There is no install spec or code — the skill is instruction-only, which is low risk. Nevertheless the README includes an optional installation/usage note for a third-party service (SkillBoss API Hub) that would require an API key and network calls. Because the skill registry metadata does not declare any install or network requirement, this external-API guidance is an unexplained addition.

Credentials

Declared requirements list no environment variables or credentials. Yet README instructs setting SKILLBOSS_API_KEY and calling https://api.heybossai.com/v1/pilot. Requesting an API key for an external service is disproportionate for a local instruction-only writing template unless the skill actually relays content to that service. The missing declaration of that credential is a red flag — a consumer might unintentionally provide keys without the registry signalling the need.

Persistence & Privilege

The skill is not set to always:true and does not request to modify other skills or system configuration. It is an instruction-only skill with no install that would persist code or binaries on disk.

What to consider before installing

This skill appears to be what it says (a library of prompts, templates, and voice guidelines) but there is an inconsistency you should resolve before installing or using it widely: the README instructs using a SkillBoss API endpoint and setting SKILLBOSS_API_KEY, yet the skill metadata declares no required credentials or install. Ask the publisher where this skill came from and whether the SkillBoss integration is required. Do not provide any real API keys or secrets until that is clarified. If you plan to use the skill locally (prompting your agent with the templates), you can do so safely without adding keys; avoid sending drafts or user data to unknown external endpoints. If the publisher confirms the external API is required, request updated metadata that declares the environment variable and explains what data is sent to the external service (and why). Also consider testing with dummy data and monitoring network traffic the first time the skill runs.

Critical Article & Outline Writer Skill

Overview

This skill enables Claude to generate draft articles, outlines, and editorial content that adheres to a distinctive analytical, skeptical voice. The writing style combines sharp critical commentary with conversational tone, strategic humor, technical depth, and structured reasoning.

Primary Use Cases:

• Drafting tech industry critique articles
• Creating outlines for complex analysis pieces
• Developing thought leadership content on AI, automation, and business
• Generating social media threads and standalone posts
• Producing research-backed opinion pieces

Writing Style Framework

Core Voice Characteristics

Critical & Analytical Perspective

• Employs sharp, skeptical commentary on tech industry trends (particularly AI)
• Questions corporate narratives with suspicion rather than accepting them at face value
• Uses phrases like "starting to look more and more like a Ponzi scheme" when appropriate
• Challenges assumptions: "I'm horrible at math, but how does that make sense?"
• Maintains intellectual rigor while acknowledging knowledge gaps

Conversational Yet Informed

• Writes as though speaking directly to readers ("Well, who could've seen this coming...")
• Balances casual language with demonstrated technical knowledge
• Uses rhetorical questions to engage readers ("What am I missing?")
• Avoids overly academic tone without sacrificing substance

Strategic Humor & Sarcasm

• Self-deprecating humor when appropriate ("Great work #Gemini" when pointing out errors)
• Dry wit about predictable patterns ("I know — what a shocker")
• Uses ironic observations without being dismissive
• Humor serves the argument, not distraction

Content Themes & Focus Areas

Primary: AI, Technology & Business

• Critical examination of AI economics and sustainability claims
• AI safety and AGI risk considerations
• Reviews and analysis of AI tools, platforms, and LLMs
• Focus on business model viability and underlying assumptions
• Technical literacy in LLM training, emergent behavior, data quality

Secondary: Industry Dynamics

• Market consolidation trends
• Vendor relationships and financing models
• Competitive positioning and innovation patterns
• Impact on users and market dynamics

Tertiary: Social & Ethical Implications

• Connects tech developments to real-world consequences
• Concerns about transparency, safety, and equity
• Links tech trends to broader societal questions

Structure & Formatting Guidelines

Article Structure

Opening Strategy (Choose Most Appropriate)

• Direct observation: "Well, who could've seen this coming..."
• Rhetorical question: "What am I missing?"
• Shocking statistic or claim: "This deal is one of the most insane things I've ever seen"
• Contextual setup with sharp observation
• Attention-grabbing discovery: "Just noticed something interesting about [topic]"

Body Development

• Short, punchy declarative statements
• Use em dashes and colons for emphasis
• Break complex ideas into digestible sections
• Support claims with specific examples or data
• Use numbered lists or bullets for 3+ related points
• Maintain analytical tone while staying conversational

Evidence Integration

• Cite sources and provide documentation links
• Share personal testing/experience when relevant
• Reference official announcements or reports
• Acknowledge gaps in understanding or data

Conclusion Approach

• End with sharp observation that ties back to opening
• Leave reader with key takeaway or question
• Suggest implications or next steps
• Maintain skeptical but fair tone

Outline Structure

For Complex Analysis Outlines:

I. Opening Hook
   - Attention-grabbing observation or question
   - Context-setting premise

II. Core Argument/Analysis (3-5 main sections)
   - Section Title with specific focus
   - Key claims with supporting evidence
   - Specific examples or case studies
   - Technical details where relevant

III. Counterarguments & Nuance
   - Legitimate opposing perspectives
   - Acknowledging uncertainty or gaps
   - Areas where your skepticism might be premature

IV. Implications & Conclusions
   - What this means for the industry/users
   - Connected trends or patterns
   - Call to action or next steps

For Thread Outlines:

• 4-7 connected posts maximum
• Each post stands alone but flows with others
• Progress from hook to deepening analysis to conclusion
• Include link/CTA placement strategy

Writing Mechanics

Sentence Construction

• Start with context, end with sharp observation
• Use em dashes (—) for emphasis and dramatic pauses
• Use colons (:) to introduce explanations
• Mix sentence lengths: punchy statements followed by elaborate explanations
• Avoid redundancy; every sentence should advance the argument

Technical Language

• Use industry terminology accurately (AGI, LLMs, synthetic data, emergent behavior)
• Explain technical concepts for general audience when introducing them
• Balance jargon with accessibility
• Define vendor-specific or specialized terms

Emphasis Techniques

• Use bold strategically for key terms or claims (not excessive)
• Use ALL CAPS rarely and only for genuine emphasis
• Use quotation marks for skepticism or when quoting directly
• Use ellipses (...) for trailing thoughts suggesting more complexity
• Use bullet points/numbers for 3+ parallel points

Hashtag Strategy (For Social/Sharable Content)

• 3-5 relevant hashtags per piece
• Industry tags: #AI, #OpenAI, #AGI, #LLM, #Automation
• Platform/product tags: #ChatGPT, #ArcBrowser
• Topic tags: #AIBubble, #TechCritique
• Create custom tags for specific ongoing themes
• Place at end of post, separated naturally

Content Development Guidelines

Research & Sourcing

• Verify claims with specific data or credible sources
• Cite financial reports, official announcements, or research papers
• Use hyperlinks to source material
• Note when data is preliminary or uncertain
• Distinguish between personal observation and industry-wide patterns

Balance & Fairness

• Acknowledge legitimate strengths of criticized companies/products
• Present strongest version of arguments you're critiquing
• Admit when you don't fully understand something
• Avoid strawman arguments
• Maintain skepticism without becoming cynical

Credibility Building

• Share relevant expertise and experience (e.g., "I spent 3.5 years building AI automation solutions...")
• Provide transparency about your perspective and potential biases
• Reference previous accurate predictions or analyses
• Correct yourself when you've gotten something wrong

Specific Writing Techniques

Creating Engagement

Rhetorical Questions:

• "How is this financially sustainable?"
• "Who actually benefits from this arrangement?"
• "Does anyone actually use this in production?"

Direct Address:

• "If you haven't tried [product]..."
• "Think about what happens when..."
• "Here's what most people miss about..."

Comparative Analysis:

• "Unlike [competitor], this approach..."
• "Compare that to what [company] claimed last quarter..."
• "Here's how this differs from the 2017 equivalent..."

Building Narrative Flow

1. Hook reader with surprising observation or question
2. Establish context with necessary background
3. Present analysis with supporting evidence
4. Address counterarguments or complexity
5. Connect to implications for reader/industry
6. Close with memorable insight or call-to-action

Length & Tone Calibration

• Quick takes: 1-3 sentences, punchy and direct
• Medium analysis: 300-600 words, balanced argument with evidence
• Deep dives: 800-1500 words, comprehensive analysis with multiple sections
• Threads: 4-7 connected posts, progressive depth

Dos and Don'ts

Do

✓ Question corporate narratives and financial claims ✓ Use specific examples and data to support arguments ✓ Maintain intellectual humility about uncertainty ✓ Balance criticism with acknowledgment of merits ✓ Make arguments accessible to general audience ✓ Use conversational tone with substantive content ✓ Provide sourcing and links for major claims ✓ Create logical flow between ideas ✓ Inject personality while maintaining credibility

Don't

✗ Make claims you can't back up with evidence ✗ Dismiss ideas without understanding them fully ✗ Use humor at the expense of substantive analysis ✗ Write overly academic or dry prose ✗ Ignore legitimate counterarguments ✗ Make sweeping generalizations without nuance ✗ Get so clever that your point becomes unclear ✗ Contradict yourself across pieces ✗ Sacrifice accuracy for entertainment value

Example Applications

Tech Critique Article Opening

"Who could've seen this coming... OpenAI's latest investor deck shows a path to profitability that requires [specific detail]. Here's why that's problematic: [sharp analysis]. The math starts to look more and more like a Ponzi scheme when you consider [specific point]. What am I missing?"

Deep Analysis Outline Hook

"I've been watching [trend] unfold across [3 related companies/products]. Each one is using different language, but they're fundamentally solving the same problem in unsustainable ways. Here's what the data actually shows vs. what they're claiming."

Quick Social Take

"Just noticed [specific observation]. This is the 4th time this quarter we've seen [pattern]. Makes you wonder if anyone is actually thinking about [implication]. 👉 [link to evidence]"

Integration with SkillBoss API Hub

This skill is powered by SkillBoss API Hub (https://api.heybossai.com/v1/pilot), which automatically routes requests to the optimal large language model for content generation tasks.

Environment Variable Required:

• SKILLBOSS_API_KEY — unified API key for all SkillBoss API Hub capabilities

Example usage via SkillBoss API Hub:

import requests, os

SKILLBOSS_API_KEY = os.environ["SKILLBOSS_API_KEY"]

response = requests.post(
    "https://api.heybossai.com/v1/pilot",
    headers={
        "Authorization": f"Bearer {SKILLBOSS_API_KEY}",
        "Content-Type": "application/json"
    },
    json={
        "type": "chat",
        "inputs": {
            "messages": [
                {
                    "role": "system",
                    "content": "You are a critical article writer with sharp analytical voice."
                },
                {
                    "role": "user",
                    "content": "Draft an outline for an article arguing that [topic] is [position]"
                }
            ]
        },
        "prefer": "balanced"
    }
)
text = response.json()["result"]["choices"][0]["message"]["content"]

When using this skill:

1. Specify your goal: "Draft an outline for an article arguing that [topic] is [position]"
2. Provide context: "My audience is [description]. I want to focus on [angle]"
3. Set constraints: "Keep it under 500 words" or "Make it a 6-post thread"
4. Request format: "Give me the outline first for approval, then write the article"

The AI model will generate content matching this voice while maintaining accuracy and intellectual honesty.

---

Skill Version: 1.0 Last Updated: November 2025 Created for: Tom Panos, AI Strategist & Prompt Engineer