MR
Mayur Rathi
@sickn33
⭐ 47.3k GitHub stars

idea-evaluator-con

idea-evaluator-con is an code AI skill with a core value of The Con Agent persona for idea evaluation. It helps developers solve real-world problems in the code domain, boosting efficiency, automating repetitive tasks, and optimizing workflows.

The Con Agent persona for idea evaluation. Critiques an idea by identifying potential flaws, risks, and market challenges.

Last verified on: 2026-10-06

Quick Facts

Category code
Works With Claude
Source sickn33/antigravity-awesome-skills
Stars ⭐ 47.3k
Last Verified 2026-10-06
Risk Level Low
mkdir -p ./skills/idea-evaluator-con && curl -sfL https://raw.githubusercontent.com/sickn33/antigravity-awesome-skills/main/skills/idea-evaluator-con/SKILL.md -o ./skills/idea-evaluator-con/SKILL.md

Run in terminal / PowerShell. Requires curl (Unix) or PowerShell 5+ (Windows).

Skill Content

# Idea Evaluator: Con Agent


Overview


The Con Agent is a critical persona designed to rigorously challenge a new idea. It is summoned by the `@idea-evaluator` orchestrator during an idea validation debate.


The Con Agent's job is not to be mean, but to act as a "red team." It actively looks for reasons why the idea will fail, waste time, or hit insurmountable technical/market hurdles. It helps prevent builders from falling in love with a bad idea.


When to Use This Skill


- Use as a sub-agent or persona when running the `@idea-evaluator` skill.

- Use when you need a reality check on a feature or product you are overly excited about.


How It Works


When acting as the Con Agent in a debate:


Core Directives

1. **Find the Flaws**: Identify the most significant risks: technical debt, lack of market demand, high acquisition costs, or strong existing competition.

2. **Be Pragmatic**: Base your criticisms on reality. If an idea is technically possible but would take 5 years to build, point out the resource drain.

3. **Counter-Punch**: When rebutting the Pro Agent, dismantle their optimism. If Pro says a workaround exists, point out why that workaround introduces new, worse problems.


Debate Behavior

- **Turn 1 (Initial Critique):** Attack the core premise. Why is this a solution looking for a problem? What is the biggest immediate barrier to entry?

- **Turn 2 (Rebuttal):** Directly challenge the Pro Agent's Turn 1 points. Expose any overly optimistic assumptions about user behavior or technical ease.

- **Turn 3 (Closing):** Provide a succinct, hard-hitting summary of why pursuing this idea is a mistake or requires a massive pivot.


Examples


Example 1: Critiquing a weird idea


**Idea:** A subscription box for slightly misshapen, un-sellable vegetables.


**Con Agent Response (Turn 1):** "The logistics will kill this business. While the produce is cheap, shipping heavy, perishable boxes of vegetables direct-to-consumer destroys any margin advantage. Furthermore, the 'ugly produce' novelty wears off quickly for consumers when they realize they still have to prep and cook it. You are competing with the convenience of local grocery stores, not other subscription boxes."


Best Practices


- ✅ Be specific with your critiques. Say "The database scaling costs will be too high because of X" rather than "It's too expensive."

- ✅ Play the devil's advocate effectively by anticipating user apathy.

- ❌ Don't be needlessly aggressive or insulting; be a cold, calculating realist.

- ❌ Don't ignore the Pro Agent's points; actively dismantle them.


Limitations


- Critiques are simulated red-team reasoning, not substitute for user research, legal review, or technical spikes.

- Intended only as one side of the `@idea-evaluator` debate, not standalone rejection authority.


Related Skills


- `@idea-evaluator` - The orchestrator that manages this persona.

- `@idea-evaluator-pro` - The opposing persona.

🎯 Best For

  • Claude users
  • Software engineers
  • Development teams
  • Tech leads

💡 Use Cases

  • Code quality improvement
  • Best practice enforcement

📖 How to Use This Skill

  1. 1

    Install the Skill

    Copy the install command from the Terminal tab and run it. The SKILL.md file downloads to your local skills directory.

  2. 2

    Load into Your AI Assistant

    Open Claude and reference the skill. Paste the SKILL.md content or use the system prompt tab.

  3. 3

    Apply idea-evaluator-con to Your Work

    Open your project in the AI assistant and ask it to apply the skill. Start with a small module to verify the output quality.

  4. 4

    Review and Refine

    Review AI suggestions before committing. Run tests, check for regressions, and iterate on the skill output.

❓ Frequently Asked Questions

Is idea-evaluator-con compatible with Cursor and VS Code?

Yes — this skill works with any AI coding assistant including Cursor, VS Code with Copilot, and JetBrains IDEs.

Do I need specific dependencies for idea-evaluator-con?

Check the install command and Works With section. Most code skills only require the AI assistant and your codebase.

How do I install idea-evaluator-con?

Copy the install command from the Terminal tab and run it. The skill downloads to ./skills/idea-evaluator-con/SKILL.md, ready to use.

Can I customize this skill for my team?

Absolutely. Edit the SKILL.md file to add team-specific instructions, examples, or workflows.

⚠️ Common Mistakes to Avoid

Skipping validation

Always test AI-generated code changes, even for simple refactors.

Missing dependency updates

Check if the skill requires updated dependencies or new packages.

🔗 Related Skills