analyze-densitylisted

Analyze text for information density, semantic repetition, and filler content. Two modes: "score" (read-only audit with per-method breakdown) and "fix" (rewrite to remove redundancy while preserving all unique information). Dispatches 7 core analysis methods (plus genre-targeted ones) as parallel subagents, then aggregates results into a composite score and actionable report. Use when asked to "check density", "find repetition", "remove AI slop", "deduplicate this text", "is this repetitive", "information density", "compress this writing", "tighten this doc", or whenever text feels bloated or AI-generated. Also use proactively when reviewing docs or PRs that contain suspiciously fluffy prose.
Max-Levitskiy/skills · ★ 0 · AI & Automation · score 70

Install: claude install-skill Max-Levitskiy/skills

# Text Density Analyzer Detect repeated meaning — not just repeated words — across documents. AI text often restates the same idea with different phrasing. This skill finds those patterns and either reports them (score mode) or fixes them (fix mode). ## Step 0: Determine target and mode If the user specified file(s), use those. Otherwise ask: ``` Which document(s) or file(s) should I analyze for information density? ``` Determine the mode from context: - **Score mode** (default): User says "check", "score", "audit", "analyze", "how dense is this" - **Fix mode**: User says "fix", "improve", "tighten", "rewrite", "compress", "remove repetition" If ambiguous, default to score mode — the user can escalate to fix after seeing results. ## Step 1: Read the target text Read all target files. If total content exceeds ~50K tokens, split into logical sections (chapters, headings, or ~2000-word chunks with overlap) and process each section independently. Track section boundaries for the final report. ## Step 2: Dispatch analysis subagents in parallel Launch all subagents in a **single message** so they run concurrently. Each subagent receives the full text (or section) and returns structured JSON. **Core methods (always run — 7):** - `agents/claim-similarity.md` — Method 1: semantic duplicate claims - `agents/compression-ratio.md` — Method 2: compressible bloat - `agents/new-information.md` — Method 3: new-info per sentence - `agents/unique-claims-ratio.md` — Method 4: unique