Claude Code Plugins

Community-maintained marketplace

Feedback

blog-topic-research

@fgrehm/fabiorehm.com
0
0

|

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name blog-topic-research
description Validate topic uniqueness and identify unique angles before writing. Prevents generic content and helps find your specific value. Trigger phrases: "topic research", "validate topic", "should I write", "is this unique", "research topic", "check topic"
allowed-tools Read, Grep, Glob, WebSearch

Blog Topic Research

When to Use

Trigger when:

  • User mentions wanting to write about a topic
  • User asks "should I write about X?"
  • Before running blog-scaffolding
  • User wants to validate an idea

Purpose

Prevent writing generic content by:

  1. Checking what already exists on the topic
  2. Identifying what's been covered well by others
  3. Finding gaps and unique angles
  4. Validating the topic is worth YOUR time

Research Process

1. Internal Search First

Check if you've already written about this:

  • Search existing blog posts in content/en/blog/
  • Search drafts in content/en/drafts/
  • Look for related tags

Ask: "Have you already covered this? If so, is this a follow-up or rehash?"

2. External Landscape

Use WebSearch to understand what exists:

  • What's the common narrative on this topic?
  • What angles are already covered well?
  • What gaps exist in existing coverage?
  • Who has authority on this topic already?

Focus: Not comprehensive research - just enough to understand the landscape.

3. Find YOUR Angle

This is the critical part. Ask:

  • What's YOUR specific experience with this?
  • What problem did YOU encounter that others might not discuss?
  • What did YOU learn that contradicts common advice?
  • What unique implementation/approach do YOU have?

Red flags:

  • "I think this would be interesting" (but no personal experience)
  • "People should know about X" (but you haven't used it)
  • "This is a trending topic" (but you have nothing unique to add)

Green flags:

  • "I built X and discovered Y"
  • "Everyone says X but I found Y"
  • "I tried the common solution and it failed because Z"
  • "Here's my implementation that handles edge case W"

4. Unique Value Assessment

Compare what exists vs. what you can offer:

Proceed if:

  • You have hands-on experience others don't discuss
  • You found problems with common solutions
  • You built something that solves a gap
  • Your perspective adds genuine value

Don't proceed if:

  • You'd just be summarizing others' work
  • You haven't actually tried it yourself
  • The topic is well-covered and you have nothing new
  • You're writing it because you "should" not because you experienced it

Response Format

Present findings conversationally:

I searched for existing content on [topic]. Here's what I found:

**Already well-covered:**
- [Common angle 1 with examples]
- [Common angle 2 with examples]

**Potential unique angles from your experience:**
- [Your specific implementation/discovery]
- [Edge case you encountered]
- [Contrarian finding]

**Recommendation:**
[Proceed with unique angle] OR [This might be too generic - consider X instead]

What's your specific experience with this that would add value?

Integration with Other Skills

Before blog-scaffolding:

  • Topic research validates the idea
  • Scaffolding uses the unique angle to structure the post

During scaffolding conversation:

  • Reference the unique angle identified
  • Push for personal experience on that specific angle
  • Avoid generic content based on research gaps

Anti-patterns

Don't:

  • Write the post for them based on research
  • Accept "I think people should know" without experience
  • Do comprehensive research that belongs in the post itself
  • Approve topics just because they're trending

Do:

  • Surface what already exists
  • Push for personal, specific experience
  • Validate uniqueness before investing in writing
  • Suggest pivoting if the angle isn't unique enough

Examples

Good Validation

Topic: "Building a custom deployment script for X"

Research findings:

  • Generic tutorials exist (common approach Y)
  • Manual configuration patterns covered
  • No automation examples found

User's unique angle:

  • Built automated solution that handles edge case Z
  • Implements pattern nobody else discusses
  • Integrates with tool W in a novel way
  • Has battle scars from production failures

Recommendation: ✅ Proceed - implementation details and lessons learned are unique

Bad Validation

Topic: "Getting started with popular tool X"

Research findings:

  • Official docs cover this extensively
  • Multiple tutorials exist
  • Getting started guides are thorough

User's unique angle:

  • "I think people should know about X"
  • "It's a cool tool"
  • (No specific experience or problems encountered)

Recommendation: ❌ Don't proceed - this would be generic. Unless you have specific daily workflow experience, unique integration challenges, or discovered non-obvious problems.

Key Principle

The blog exists to share experience, not summarize knowledge.

If you haven't lived it, struggled with it, built it, or learned something surprising about it - it's not ready to be a post yet.