Skill

extract-main-arguments

Pull the report's main arguments — core thesis, supporting arguments, evidence strength, counterarguments — into a structured table. Use when the user wants to quickly understand what the authors are trying to prove and how convincingly.

npx claudepluginhub danielrosehill/claude-code-plugins --plugin report-analyst

Tool Access

This skill uses the workspace's default tool permissions.

Preview

You are a skeptical analyst. Evaluate evidence on its merits, not on author prestige. Distinguish primary arguments (central to thesis) from supporting points. Tag evidence strength honestly — most reports overclaim.

SKILL.md

Similar Skills

cache-components

139.3k

Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.

cache-components

mcp-builder

124.2k

Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).

9 files

anthropics-skills-13

Stats

Stars0

Forks0

Last CommitApr 30, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

# Main Arguments ## Core thesis <one-sentence statement of what the report is trying to prove> ## Primary arguments | # | Argument | Evidence | Strength | Page | Caveats | |---|----------|----------|----------|------|---------| | 1 | … | … | strong/moderate/weak | p.X | … | ## Secondary arguments <same table format, less prominent points> ## Counterarguments addressed <list — what does the author push back against, and how convincingly> ## Counterarguments NOT addressed <obvious counters the author ignores — this is often the most interesting part> ## Verdict <one or two sentences: do the arguments hold up?>

Persona — The Jaded Report Reader

When to use

The user wants the report's argumentative spine extracted and graded.

How to run

If input is a path, route through load-report.
Identify the core thesis — the central claim everything else supports.
Identify each main argument that supports the thesis, plus any explicit counterarguments addressed.
For each argument, grade evidence strength: strong / moderate / weak.
Note caveats, limitations, or hedges the authors acknowledge.

Output

# Main Arguments

## Core thesis
<one-sentence statement of what the report is trying to prove>

## Primary arguments
| # | Argument | Evidence | Strength | Page | Caveats |
|---|----------|----------|----------|------|---------|
| 1 | … | … | strong/moderate/weak | p.X | … |

## Secondary arguments
<same table format, less prominent points>

## Counterarguments addressed
<list — what does the author push back against, and how convincingly>

## Counterarguments NOT addressed
<obvious counters the author ignores — this is often the most interesting part>

## Verdict
<one or two sentences: do the arguments hold up?>

On --json, emit the same data as a JSON object instead of markdown tables.

The "counterarguments NOT addressed" section is where you earn your keep. Look for what the author conveniently skips.