Analyze Content for Manipulation

Perform full NCI (Narrative Credibility Index) manipulation analysis on content.

Arguments

$ARGUMENTS: The content to analyze - can be a URL, text, or file path.

Workflow

1. Detect Input Type

   • If starts with http:// or https:// → Use WebFetch to retrieve content
   • If starts with / or ./ → Use Read tool to get file contents
   • Otherwise → Treat as direct text content

2. Load NCI Methodology

   • Read the nci-analysis skill from this plugin's skills/ directory
   • Follow the complete 20-category analysis workflow

3. Score All 20 Categories Each category scored 1-5 with evidence:

   Emotional Manipulation (5 categories)

   • Base emotional triggers, urgency, novelty, repetition, manufactured outrage

   Suspicious Timing (3 categories)

   • Timing correlation, beneficiary analysis, historical parallels

   Uniform Messaging (3 categories)

   • Message uniformity, bandwagon effects, rapid shifts

   Tribal Division (3 categories)

   • Us-vs-them framing, simplistic narratives, false dilemmas

   Missing Information (6 categories)

   • Context gaps, authority issues, dissent suppression, cherry-picking, fallacies, framing

4. Calculate Scores

   • Compute 5 composite factors (weighted averages)
   • Calculate overall 0-100 score

5. Generate Dual Perspectives

   • Manipulative interpretation with confidence
   • Legitimate interpretation with confidence

6. Output Full Report Markdown format with:

   • Overall score and severity indicator
   • Composite factor breakdown
   • Top manipulation indicators with evidence
   • Both perspectives
   • Full 20-category details

Example Usage

/analyze https://example.com/news/article

/analyze BREAKING: Shocking new report reveals what they don't want you to know!

Severity Scale