agent-skills / rules
Cursor rule28 starsChanged 3 months agoMITView on GitHub
Load project memory from MASTER_CONTEXT.md and keep it updated after substantive work.
- Reads credentials
Where files are headingNew Cursor rule files have more rules about planning and tracking tasks, and fewer about code style.
Of the Cursor rule files we read, the 4,181 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- shared/CLAUDE.mdCLAUDE.md
- analyze-videoSkill
- change-voiceSkill
- chatgpt-image-adSkill
- claymation-adSkill
- clone-image-adSkill
- clone-video-adSkill
- generate-youtube-thumbnailSkill
- image-to-motionSkill
- nano-banana-image-adSkill
- novoads-apiSkill
- novoads-updateSkill
- pixar-adSkill
- spy-competitor-adsSkill
The file
We split this project-context.mdc into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
If .env or MASTER_CONTEXT.md do not exist, tell the user to run ./scripts/setup.sh.
- Very specific: line 11
After significant changes (new workflows, verified Novoads behaviors, brand updates), append a short dated entry to the Changelog section …
≈ “Common” is an estimate: at least 1 in 5 of the 6,668 Cursor rule files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
---
description: Load project memory from MASTER_CONTEXT.md and keep it updated after substantive work.
alwaysApply: true
---
# Project context
- If `.env` or `MASTER_CONTEXT.md` do not exist, tell the user to run `./scripts/setup.sh`.
- At the **start** of substantive work in this repo, read **MASTER_CONTEXT.md** at the repository root for brand voice, default product, and API learnings. It carries **no prices** — that is deliberate.
- The first time the user asks to generate something and `MASTER_CONTEXT.md` is missing a field that request needs (default product, brand voice), ask for it then — once — and write the answer back so no future session asks again. A setup-only session asks for nothing and reports almost nothing: close with the setup-close template in `AGENTS.md` ("First-time setup" — a status line, "What you can ask for now", the product-photos line) and keep git mechanics and sync details out unless they block an ask. Never write a credit number into it.
- After **significant** changes (new workflows, verified Novoads behaviors, brand updates), append a short **dated** entry to the Changelog section in **MASTER_CONTEXT.md**.
- For Novoads HTTP usage, prompts, and polling, follow **`.cursor/skills/novoads-api/SKILL.md`** and linked **reference.md**.
- **Session work-product goes in a gitignored home** — `generated/` (renders), `outputs/<job>/` (video downloads + edit workfiles), `prompts/` (composed prompts), `iterations/` (clone rounds), `logs/`. Never invent a new top-level directory; leave `git status` as clean as you found it. Stated in full in `AGENTS.md` ("Every session").
- **Never quote a credit cost from memory.** Prices come from a live `POST /v1/estimates` in the current session, shown and approved before anything is generated. For a video where the model speaks, the spoken line is approved separately, first.
- Log every generation call to `logs/novoads-api.jsonl` — observability only, never a pricing source.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.