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/exercise-syntaxgit clone --depth 1 https://github.com/AI-Learning-Gems/AI-Learning-Gems.github.ioWhat 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 | $0.02068 | $0.02068 |
| Opus 5 | $0.01034 | $0.01034 |
| Sonnet 5 | $0.00414 | $0.00414 |
| Haiku 4.5 | $0.00207 | $0.00207 |
Grade A, and why
exercise-syntax 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 2d 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 — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Exercise Syntax Rules
Exact syntax rules for the interactive exercises Quarto extension (_extensions/exercises/). These rules are non-negotiable: exercises that violate them will silently break (render as raw text, fail to submit, or crash with JS errors).
The extension uses a Lua filter that walks the Pandoc AST. Pandoc parses the markdown before the filter runs, so the markdown must produce the exact AST structures the filter expects.
General Rules (All Exercise Types)
-
Blank lines before and after every exercise block. The outer
::: {.exercise-*}and closing:::must have blank lines above and below them. -
Options MUST use bullet list syntax (
- Option text), on consecutive lines with NO blank lines between items. Pandoc parses- itemas a BulletList node; the Lua filter searches for BulletList. If you use standalone paragraphs, numbered paragraphs, or lettered paragraphs (likeA. textwith blank lines), Pandoc will parse them as separate Para nodes and the filter will not find the options.
# CORRECT — consecutive bullet items, no blank lines between:
- Option text one
- Option text two
- Option text three
# WRONG — standalone paragraphs (Pandoc sees Para, not BulletList):
A. Option text one
B. Option text two
C. Option text three
# WRONG — numbered list (Pandoc sees OrderedList, not BulletList):
1. Option text one
2. Option text two
3. Option text three
- Do NOT prefix options with letter labels in any format:
A),B),A.,B.,[A],[B], etc. The Lua filter auto-assigns labels A, B, C, D based on list order. These prefixes break in different ways:A)triggers Pandoc's ordered list parser producing<ol type="A">, while[A]renders as literal bracket text inside the option card, duplicating the auto-assigned label.
# CORRECT — plain text, labels auto-assigned:
- An agent uses a larger language model
- An agent has a decision procedure that controls a loop
- An agent always uses multiple tools simultaneously
# WRONG — "A)" triggers Pandoc's ordered list parser:
- A) An agent uses a larger language model
- B) An agent has a decision procedure
# ALSO WRONG — "[A]" renders as duplicate label text:
- [A] An agent uses a larger language model
- [B] An agent has a decision procedure
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.
- 2d ago First seen · 208 lines · 2,068 tokens per session scan A e9011d479440
exercise-syntax 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 2,068 tokens to every session, about $0.0103 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.
typescript
Changes to these high-fan-out internals can affect every message, delta, element, or rerun. Keep work in them minimal, and benchmark changes with representative stress-test apps.
coolify-ai-docs
Master reference to all Coolify AI documentation in .ai/ directory.
python_lib
Tips and guidelines specific to the development of the Streamlit Python library, not applicable to scripts and e2e tests.
specs
This directory contains product and tech specs for Streamlit features.