Claude Code Plugins

Community-maintained marketplace

Feedback

proof-points

@nphillips/2025-12-24-proof-points-skill
0
0

Extract proof points from portfolio case studies and generate punchy scripts with storyboards. Use when user wants to create impactful narratives about their work, develop case study content, write portfolio scripts, identify what makes their project valuable, or pair achievements with visuals. Triggers on requests like "help me explain this project," "what did I improve here," "create a script for this case study," or "extract the value from this work.

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 proof-points
description Extract proof points from portfolio case studies and generate punchy scripts with storyboards. Use when user wants to create impactful narratives about their work, develop case study content, write portfolio scripts, identify what makes their project valuable, or pair achievements with visuals. Triggers on requests like "help me explain this project," "what did I improve here," "create a script for this case study," or "extract the value from this work."

Proof Points

Extract specific, verifiable improvements from projects and assemble them into impactful narratives.

Workflow

1. Gather Input

Accept either:

  • Visual evidence: Screenshots, flipbook demos, before/after comparisons
  • Description: User explains what they did, the problem, the solution

If visual evidence provided, analyze it first. If description only, ask clarifying questions about specific improvements made.

2. Extract Draft Proof Points

Identify 2-5 proof points. Each proof point must be:

  • Specific: "horizontal nav truncates on tablet" not "it wasn't responsive"
  • Verifiable: Can be demonstrated with evidence
  • Client-value-focused: Frame as business outcome, not just technical fix

Present draft proof points to user:

"Here's what I'm seeing as the key improvements. Did I get this right? What am I missing?"

3. Refine Collaboratively

Apply these principles when refining:

  • Lead with problem, not process - Nobody cares about "conducted research" until they feel the pain
  • Technical specifics = credibility - Concrete details beat vague claims
  • Frame fixes as client outcomes - "Users couldn't add restaurants because the integration was broken" > "I fixed the script"
  • Contrast creates impact - Before/after in the same breath

Ask about anything not visible in screenshots:

"Are there improvements that aren't immediately visible here?"

4. Brainstorm Visuals

For each proof point, propose the most impactful visual:

  • Before/after comparison
  • Screenshot of the specific problem
  • Diagram showing the fix
  • Demo/interaction recording

User will create/capture these later - just describe what would work best.

5. Assemble Output

Generate a Markdown file with:

## Script

[Hook - attention-grabber, 1-2 sentences]

[Problem - the pain point, what was broken]

[Proof Points - 2-4 specific improvements with contrast]

[Outcome - client value delivered]

## Storyboard

| Text | Image |
|------|-------|
| [Script segment] | [Description of ideal visual] |
| ... | ... |

Terminology

Term Meaning
Hook Opening attention-grabber (first 1-3 seconds of video, or first sentence)
Proof points Specific, verifiable improvements - the core deliverable
Outcome Client value delivered - the "so what"

Output Flexibility

The script works for multiple formats:

  • YouTube Shorts (9:16 video with narration)
  • Case study hero slideshow (16:9 or square)
  • Written intro paragraph with supporting images
  • Interactive comparison demos

The storyboard's "Image" column describes ideal visuals - user decides execution format.