ai-asset-pricing: Skill for Claude Code

Style Check Skill

When this skill is invoked, analyze the provided text for violations of the academic writing standards in .claude/rules/academic-writing.md.

Examples

• /style-check -- check the current .tex file
• /style-check introduction -- check only the introduction section
• /style-check projects/my_paper/latex/main.tex -- check a specific file

Input

The user provides a .tex file path or text block to analyze.

Workflow

1. Read .claude/rules/academic-writing.md for the full rule set
2. Read the project's CLAUDE.md for project-specific terminology
3. Read the target text (from file path or section markers)
4. Run all 13 analysis categories below
5. Check citations against the project's .bib file
6. Produce the structured report

Analysis Categories

1. Banned Word Scan

Check for all terms in academic-writing.md Section 1 and flag each occurrence:

• delve, crucial, comprehensive, multifaceted, utilize, leverage, facilitate
• endeavor, paramount, myriad, plethora, noteworthy
• "it is important to note", "it should be noted", "in this regard"
• "the fact that", "in order to", "due to the fact that"
• "a number of", "the vast majority of", "at the present time"
• "in the context of", "with respect to", "in terms of", "as such"
• All AI-marker words: underscores, showcasing, pivotal, intricate, meticulous, illuminate, unveil, bolster, realm, landscape, foster, encompass, aligns with, shed light on, profound, grappling, etc.

2. Throat-Clearing Detection

Flag openings that don't state a concrete finding:

• "The literature has long..."
• "Financial economists have wondered..."
• "An important question..."
• "Recent years have witnessed..."
• "It is well known that..."
• "There has been growing interest in..."
• "A growing body of literature..."

3. Passive Voice Check

Flag passive constructions:

• "it is shown that" -> "we show that"
• "it was found that" -> "we find that"
• "can be seen" -> "shows"
• "has been documented" -> "[Author] documents"