Lauren Poteto Rules skill
A skill of working rules for coding agents: diagnose from code, verify in the running product, turn corrections into constraints.
- 98 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- The skill holds instructions only, but they make the agent run your app and checks with your permissions.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/lauren-poteto-rulesDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skill holds instructions only, but they make the agent run your app and checks with your permissions.
This is third-party code. Review the repository files before installing.
What it does
A single `SKILL.md` of engineering rules adapted from Lauren Tan's talks on working with coding agents. The agent reads the affected code before naming a cause, then verifies a fix in the running product: a passing build or type check does not count as proof. The second half of the rules is about the environment. A recurring agent mistake is to be closed with a type, a lint rule, a CI check or a documentation entry. Autonomy grows gradually, while results stay reviewable and reproducible. The repository includes `EXAMPLES.md` with typical failure modes.
Who it is for. Developers whose agent confidently explains a bug from a guess or reports a task done without running the app.
Good fit when
- The agent named a cause without opening the relevant code, and you want line references.
- A fix passed the build, but nobody exercised the behaviour in the browser or CLI.
- The same review comment keeps coming back from task to task.
- Part of the work goes to sub-agents, and you need a report listing the checks actually run.
Not a fit when
- Fixing a typo or one documentation line: a plain request is enough.
- You need ready build and test commands for your stack: the skill has none.
- You expect the app to start automatically: the skill only tells the agent to find the repository's run method.
Example request
Use lauren-poteto-rules: find why the export fails, show the code and reproduction steps, then verify the fix in the running app.Limitations
The skill is plain text instructions with no code or automated checks. Results depend on the model and on how well the repository documents running the app. The README targets Grok Bot and Grok CLI, and in Cursor the rules are pasted into an Agent Requested rule by hand. It is an independent community adaptation, not an official release from the speaker. The skill targets bounded, verifiable work and does not replace review.
How to disable. Delete the `lauren-poteto-rules` folder from the agent skills directory, or the rule from `.cursor/rules`.
FAQ
How do I invoke the skill in Claude Code?
With `/lauren-poteto-rules` after cloning into `~/.claude/skills`. For a single project, place the folder in the repository's `.claude/skills`.
How do I invoke it in Codex?
Write `$lauren-poteto-rules` in a new session. Clone the folder into `~/.agents/skills` or the project's `.agents/skills`.
How do I add it to Cursor?
Create an Agent Requested rule and paste the text of `SKILL.md` into it. Alternatively, add the instructions to the root `AGENTS.md`.
How can I check the skill works?
Run the same tasks in fresh sessions with and without it, then compare actions and results. The agent's own confidence does not count.
What is the licence?
MIT.
Discussion
No questions or reviews yet.
Related
SkillsA library of 30-plus instruction skills for coding agents: planning, building, testing, hallucination guards and shipping
Two skills for coding agents: wtf audits a project before shipping, better-codex keeps the agent disciplined while working on code
Agent rules from programming books
AI agents Rules / Skills from Programming Books
Skills and AGENTS.md rules distilled from classic books: Clean Code, Refactoring, DDD, Clean Architecture, DDIA and more
14 skills that find the gaps in a task before the code: interviews, plans, assumption tests and proof that a bug fix works
19 skills for UI, code and copy: the agent decides from project evidence and verifies results rather than making a template.
A skill that runs a full refactor so agents read code more easily: it finds hidden links and needless dependencies.
README in short
The README covers seven principles, installation for Grok Bot, Grok CLI, Codex, Claude Code and Cursor, a request template, and a way to judge the skill by the agent's actions.