humanize-readme
Humanize README
Reads the current README.md, audits it for AI slop patterns, then rewrites it in a direct, honest, human voice.
Workflow
Step 1 — Read the README
cat README.md
Also check what the project actually is (to rewrite with specifics, not generics):
cat package.json 2>/dev/null | head -20
cat pyproject.toml 2>/dev/null | head -15
ls src/ 2>/dev/null | head -10
Step 2 — Audit for slop
Read references/slop-patterns.md for the full list. Flag these in the README:
High-signal slop patterns:
- Banned buzzwords:
seamlessly,robust,scalable,leverage,cutting-edge,comprehensive,empower,intuitive,powerful,game-changer - 2026-era hedging/filler:
"it's worth noting","let's explore","this is designed to","makes it easy to","enables developers to","ensures","has been designed" - Generic openers starting with "In today's...", "This powerful tool...", "This repository aims to..."
- Feature lists with empty adjectives: "Blazing fast", "Enterprise-grade", "Intuitive API"
- Suspiciously polished completeness with no honest gaps
- Conclusions that philosophize about the project
- Uniform sentence length — every sentence the same rhythm (burstiness check)
- No personal voice — no "why it exists", no honest limitations, no tradeoffs mentioned
Step 3 — Rewrite
Rewrite the README applying these rules:
Voice:
- Write like you'd explain the project over coffee — direct, specific, a bit casual
- Use the actual tech names (not "modern technologies" or "industry-standard tools")
- Keep code blocks, commands, and links exactly as they are
- Preserve the structure (sections) but rewrite the prose
For each section:
- Project description / intro — one or two sentences: what it does, why it exists. No superlatives.
- Features — remove adjectives, add specifics. Not "Fast" → state the actual number if known, otherwise just name the feature plainly
- Installation / Usage — keep as-is if already good; strip any "Welcome to the getting started..." filler
- Why this project — if it exists and sounds generic, rewrite with a real reason or remove it
- Contributing / Closing — strip "star this repo", philosophy, or over-long contributing guides
Tone rules:
- Honest about gaps: "Not tested on Windows", "Still experimental", "Works on my machine"
- Varied sentence lengths — short and long, not all the same
- No emoji unless they were already there (and even then, fewer)
- No exclamation marks unless genuinely warranted
Do NOT:
- Add content that wasn't there — only rewrite existing content
- Change technical accuracy — keep the same claims, just strip the fluff
- Make it curt to the point of being unhelpful — clarity > brevity
- Touch badge rows (
[](#)lines) — CI badges, version badges, license badges stay exactly as-is
Step 4 — Output
Output the full rewritten README in a single fenced markdown code block so it can be copied directly.
Before the block, briefly note what you changed (2-3 bullet points max).
For full banned phrase list and before/after examples: references/slop-patterns.md
More from b4r7x/agent-skills
react-design-patterns
Use when choosing a React component pattern — custom hooks, control props, compound components, headless components, render props, container/presentational, or other architectural patterns. Includes 13 patterns with decision guide and 2025 popularity ranking.
26human-commit
Generates human-like git commit messages based on staged or unstaged changes. Reads git diff, analyzes what changed, and outputs 3 natural commit message options that sound like they were written by a developer — not AI. This skill should be used when the user wants a commit message, asks "what should I write for commit", "generate commit message", "human like commit", "wiadomość do commita", or just asks for help committing.
24improve-prompt
Transforms a rough, unpolished prompt idea into a precise, structured AI coding prompt. Automatically researches the current project context (stack, file structure, conventions, git history) before generating. This skill should be used when the user provides a vague or "dirty" prompt idea and asks to refine, improve, or rewrite it — e.g. "improve this prompt", "refine my prompt", "ulepszony prompt", "dopracuj prompt", or simply describes what they want done in rough terms.
23react-anti-patterns
Use when reviewing React code — especially AI-generated code — to catch common anti-patterns. Covers 18 anti-patterns with detection difficulty, including stale closures, state mutation, useEffect abuse, and boolean explosion.
21deep-plan
Takes a rough, unpolished prompt idea and autonomously turns it into an implementation plan. Researches the project deeply, asks clarifying questions, generates a precise internal prompt, then executes it to produce a structured plan with todos. Designed for plan mode. Use when the user gives a vague feature request, rough idea, or "dirty" prompt and wants a ready-to-execute implementation plan — e.g. "plan this", "deep plan", "turn this into a plan", "zaplanuj to", "zrób plan".
19react-useeffect
Use when writing, reviewing, or debugging useEffect in React. Covers the 6 valid use cases, 7 anti-patterns, dependency pitfalls, and the decision tree for whether you actually need an effect.
18