AI Skill StudioAI Skill Studio
Skill Library
Presets:

Target Standard & File Format

AI Agent Rules & Skills
Context Shields & Privacy (Ignore Files)
Integrations & Machine Documentation
AI Governance & Multi-Agent Planning Suite

.cursorignore Shield

Prevent Cursor AI from indexing secrets, build caches, and bulky lockfiles to save context window tokens:

Full Protection
Tokens Saved / Turn
~190k tokens
Secret Leak Guards
2 Active Guards
Shield Categories
8 of 8 Enabled

Indexing Boundary Categories

Toggle specific categories to mask secrets and exclude heavy artifacts:

Custom Rules & Negate Exceptions

Append repo-specific file globs or re-include allowed files with negation patterns (!path):

DevOps & Tooling · Cursor

Kubernetes & Helm Charts .cursorignore

A .cursorignore for Kubernetes & Helm Charts that keeps build output, large files and secrets out of Cursor's index (best effort).

Path
.cursorignore
Loads
Indexing and AI features
Format
gitignore syntax
Used by
Cursor
All .cursorignore presets and how .cursorignore works

What this file tells the agent

This .cursorignore keeps Kubernetes & Helm Charts build output, large data files and environment secrets out of Cursor's index and AI features, so context goes to your source code. It is best effort: commands the agent runs in the terminal can still read ignored files, so keep real secrets out of the repo too.

  • Always specify explicit resources.requests and resources.limits (both CPU and memory) on every container.
  • Enforce strict security contexts: readOnlyRootFilesystem: true, runAsNonRoot: true, and allowPrivilegeEscalation: false.
  • Mandate both livenessProbe and readinessProbe with appropriate initialDelaySeconds on deployment workloads.
  • Validate Helm template indentation using nindent filters (e.g. toYaml . | nindent 8) to eliminate YAML formatting corruption.
  • Configure PodDisruptionBudgets (PDB) and topology spread constraints for multi-node high availability.

Code to avoid, and what to write instead

The rule steers the agent away from the left and toward the right.

avoid
# Dangerous: No resource constraints, root container, missing probes
apiVersion: apps/v1
kind: Deployment
metadata:
  name: bad-service
spec:
  template:
    spec:
      containers:
        - name: app
          image: myapp:latest # Mutable tag and unconstrained limits!
write instead
apiVersion: apps/v1
kind: Deployment
metadata:
  name: api-service
  labels:
    app.kubernetes.io/name: api-service
spec:
  replicas: 3
  template:
    spec:
      securityContext:
        runAsNonRoot: true
        runAsUser: 10001
      containers:
        - name: api
          image: registry.example.com/api:v1.4.0
          securityContext:
            allowPrivilegeEscalation: false
            readOnlyRootFilesystem: true
          resources:
            requests:
              cpu: 100m
              memory: 128Mi
            limits:
              cpu: 500m
              memory: 512Mi
          readinessProbe:
            httpGet:
              path: /healthz
              port: 8080
            initialDelaySeconds: 5

Install it in one command

  1. 01

    Go to your project root

    The file goes at .cursorignore, relative to the root.

  2. 02

    Download the file

    Run the command on the right. It downloads this preset and writes the file.

  3. 03

    Make sure it gets used

    Cursor applies .cursorignore automatically. Matching files leave the index and AI features, but commands the agent runs in the terminal can still read them.

$ curl -fsSL "https://www.devscratchpad.tech/api/raw/cursorignore/kubernetes-helm" -o ".cursorignore"

writes .cursorignore

path

.cursorignore

To write the file from your own code instead, use Read your code or Convert Legacy Rules in the editor above.

Questions

How is .cursorignore different from .gitignore?
.gitignore decides what git tracks. .cursorignore keeps files out of Cursor's codebase index and AI features; Cursor already skips files in .gitignore when it indexes.
Does .cursorignore reduce wasted context?
Yes. Ignored files are not indexed or pulled into requests, so build output and large data files do not crowd out your source code.