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.
git clone --depth 1 https://github.com/zenbase-ai/llmlWrote 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/zenbase-ai/llml/py)<a href="https://agentmods.dev/rules/zenbase-ai/llml/py"><img src="https://agentmods.dev/badge/rules/zenbase-ai/llml/py/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/zenbase-ai/llml/py"><img src="https://agentmods.dev/badge/rules/zenbase-ai/llml/py.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00000 | $0.00843 |
| Opus 5.5 | $0.00000 | $0.00337 |
| Sonnet 5.5 | $0.00000 | $0.00169 |
| Haiku 4.5 | $0.00000 | $0.00084 |
Grade A, and why
py 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 1mo 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Python Coding Rules for LLML Project
Project Structure and Management
- Configuration: All project metadata, dependencies, and tool configurations must be managed in
pyproject.toml. - Dependencies:
- Runtime dependencies should be added to the
[project.dependencies]section. - Development dependencies must be added to the
[project.optional-dependencies]section under thedevgroup. - Use
uvfor managing the virtual environment and installing dependencies.
- Runtime dependencies should be added to the
- Source Code: All main application logic must reside in the
py/src/directory. - Tests: All tests must be placed in the
py/tests/directory and follow the naming conventiontest_*.py.
Coding Style and Formatting
- Linter/Formatter:
ruffis the designated tool for all linting and formatting. All code must be compliant with the rules defined inpyproject.toml. - Line Length: The maximum line length is
88characters. - Quotes: Use double quotes (
") for all strings. Single quotes are not permitted. - Indentation: Use 4 spaces for indentation. Tabs are not allowed.
- Docstrings:
- All public modules, functions, and methods must have a docstring.
- Use triple-double quotes (
"""Docstring goes here""") for docstrings. - Function docstrings should clearly describe the function's purpose, arguments, and return value.
Typing and Type Safety
- Type Hinting: All function signatures, including arguments and return values, must have type hints from Python's
typingmodule. - Runtime Type Checking: The
@beartypedecorator must be applied to all functions to enforce runtime type safety. - Type Aliases: Use the
talias for thetypingmodule (e.g.,import typing as t).
Function and API Design
- Core Logic: The main logic is centered around the
llmlfunction, which is designed to be recursive. When adding new features, maintain this recursive pattern. - Immutability: The
llmlfunction should be pure and not modify its inputs. It should return a new string as the result. - Keyword Arguments: When calling functions, prefer keyword arguments for clarity, especially for functions with multiple parameters.
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.
- 1mo ago First seen · 65 lines · 0 tokens per session scan A 85d7dad04b95
py is a cursor rule published in the GitHub repository zenbase-ai/llml (73 stars, last pushed 1y ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 843 tokens. 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 2026-09-01.
Other cursor rules, from other repositories
python_tests
We use the unit tests to cover internal behavior that can work without the web / backend counterpart. We aim for 95%+ unit test coverage of our Python code in lib/streamlit.
python-llm-ml-workflow-cursorrules-prompt-file
Cursor rules for Python LLM & ML development with workflow integration.
python
Python best practices and patterns for modern software development with Flask and SQLite.
python-containerization-cursorrules-prompt-file
Cursor rules for Python development with containerization integration.
python-django-general
General coding guidance for Python and Django projects, including naming, formatting, modular apps, built-in tools, and maintainable structure. Python is a programming language, and Django is a Python web framework.
python-general-coding-standards
A set of general standards for Python projects, including type hints, input checking, code style, security, web requests, background work, and tests. It also covers FastAPI-specific tasks such as CORS and authentication.