Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/ai-learning-gems/ai-learning-gems.github.io/writing-stylegit clone --depth 1 https://github.com/AI-Learning-Gems/AI-Learning-Gems.github.ioWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/ai-learning-gems/ai-learning-gems.github.io/writing-style)<a href="https://agentmods.dev/rules/ai-learning-gems/ai-learning-gems.github.io/writing-style"><img src="https://agentmods.dev/badge/rules/ai-learning-gems/ai-learning-gems.github.io/writing-style.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.09039 | $0.09039 |
| Opus 5 | $0.04519 | $0.04519 |
| Sonnet 5 | $0.01808 | $0.01808 |
| Haiku 4.5 | $0.00904 | $0.00904 |
Grade A, and why
writing-style scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 541 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Writing Style Rules
Shared writing style rules for all textbook chapter workflows (research, write, edit, update).
Target reader: Someone with strong reading comprehension and technical background who is new to this specific topic. Write with depth and precision, not oversimplification.
Engaging Writing (Make It Unputdownable)
The best technical textbooks (MacKay, Sutton & Barto, Feynman) are engaging because they use techniques from great non-fiction writing. Apply these:
Conversational Tone (Like Explaining to a Smart Friend)
- Write as if you're having a conversation with the reader
- Address the reader as "you" directly
- Show enthusiasm: "This is the beautiful part." "Here's where it gets interesting."
- Acknowledge when something is confusing: "This trips up everyone at first."
- Be intellectually honest about uncertainty: "We don't fully understand why, but..."
Sentence Rhythm and Variety (Gary Provost's Principle)
Vary sentence length deliberately. Short sentences punch. Long sentences build momentum and carry the reader through complex ideas with the energy of a crescendo.
- Short sentences for emphasis: "That's wrong." "Here's why." "This matters."
- Medium sentences for flow: Carry the main explanation forward.
- Long sentences for buildup: Where appropriate, build momentum.
- Deliberate fragments for punch: "Music." "Exactly." "Finally."
Micro-Surprises and Dopamine Hits
Keep readers engaged with small rewards throughout:
| Technique | Example |
|---|---|
| Surprising facts | "You might expect X, but actually Y" |
| Rhetorical questions | "But wait — how can that be?" |
| Vivid analogies | "Like untangling earbuds — tedious, but oddly revealing" |
| Pattern interrupts | Start a section with something unexpected |
| Open loops | Tease what's coming: "We'll see why this matters in Section 3" |
| Enthusiasm markers | "This is where it gets good." |
Motivation Before Formalism
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 541 lines · 9,039 tokens per session scan A 6e38643a32df
writing-style is a cursor rule published in the GitHub repository AI-Learning-Gems/AI-Learning-Gems.github.io (22 stars, last pushed 2mo ago), licensed MIT. It adds 9,039 tokens to every session, about $0.0452 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.
family-instance-domain-actions
Family instance domain action implementation patterns.