agent-stack-template / rules
Cursor rule31 starsChanged 3 months agoMITView on GitHub
Server Action, service and tenancy contract
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 4,183 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 repository20 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .agents/skills/react-hook-form/AGENTS.mdAGENTS.md
- .agents/skills/vercel-react-best-practices/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/git-flow.mdcCursor rule
- .cursor/rules/performance.mdcCursor rule
- .cursor/rules/prisma-schema.mdcCursor rule
- .cursor/rules/project.mdcCursor rule
- .cursor/rules/tdd-domain.mdcCursor rule
- .cursor/rules/ui-design-system.mdcCursor rule
- better-auth-best-practicesSkill
- better-auth-security-best-practicesSkill
- changelog-automationSkill
- frontend-designSkill
- interface-designSkill
- marketing-psychologySkill
- organization-best-practicesSkill
- react-hook-formSkill
- shadcnSkill
- vercel-react-best-practicesSkill
The file
We split this server-actions.mdc into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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 14
Always build them with defineAction or defineWorkspaceAction (src/lib/action.ts).
- Very specific: line 49
Start independent I/O together (Promise.all).
- Distinctive: line 52
No cross-request caching of tenant data.
≈ “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: Server Action, service and tenancy contract
globs: src/features/**/actions/**/*.ts,src/features/**/services/**/*.ts,src/app/api/**/*.ts
alwaysApply: false
---
# Server boundary
A Server Action is a public HTTP endpoint. Middleware and layout guards do not
protect it — it can be invoked directly with any payload.
## Actions
Always build them with `defineAction` or `defineWorkspaceAction`
(`src/lib/action.ts`). They run, in order: session check → Zod parse →
membership lookup → handler → error mapping.
```ts
"use server";
export const renameProjectAction = defineWorkspaceAction({
input: renameProjectSchema, // must contain workspaceId
handler: async ({ input, ctx }) => {
assertCanWrite(ctx.role); // role rules stay explicit
await renameProject({ ...input, workspaceId: ctx.workspaceId });
revalidatePath("/projects");
},
});
```
- Scope every write with `ctx.workspaceId` and `ctx.userId`, never with an id
taken from the payload.
- Actions return `ActionResult`; they do not throw at the UI and never leak a
stack trace.
- Map a domain `code` to specific copy through the `errors` option instead of
branching on `instanceof` in the handler.
## Services
- The only layer that calls Prisma.
- Load what the domain needs, call the pure rule, persist the result. Never
re-implement a rule that belongs in `domain/`.
- Scope reads *and* writes by `workspaceId` — that filter, not the action
wrapper alone, is what makes a cross-tenant id return "not found".
- `import "server-only"` at the top.
## Performance
- Start independent I/O together (`Promise.all`); do not chain awaits that have
no dependency.
- Wrap per-request reads that several components need in `React.cache`.
- No cross-request caching of tenant data. A revoked membership must take
effect on the next request.
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.

