Cloudflare Zero Trust logo

Cloudflare Zero Trust

CommunityPopular
BagelHole
cloudflare-zero-trust

Protect internal apps with Cloudflare Access, device posture, and Zero Trust policies.

Overview

PublisherBagelHole
RepositoryDevOps-Security-Agent-Skills
Skill namecloudflare-zero-trust
Stars
1.1K
Forks
161
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by BagelHole on GitHub. Read the source before you install it.

Installation

Install the Cloudflare Zero Trust AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/BagelHole/DevOps-Security-Agent-Skills/tree/main/infrastructure/cloudflare/cloudflare-zero-trust
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/BagelHole/DevOps-Security-Agent-Skills.git /tmp/DevOps-Security-Agent-Skills
mkdir -p .claude/skills
cp -r /tmp/DevOps-Security-Agent-Skills/infrastructure/cloudflare/cloudflare-zero-trust .claude/skills/cloudflare-zero-trust
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Cloudflare Zero Trust in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Cloudflare Zero Trust on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Cloudflare Zero Trust is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Cloudflare Zero Trust

Secure access to internal services without VPNs using Cloudflare's Zero Trust platform (Access, Tunnel, Gateway, and WARP).

When to Use

  • Replacing VPN access to internal web applications, SSH, or RDP.
  • Enforcing identity-aware access policies on internal tools (dashboards, admin panels).
  • Exposing on-premises or private-network services securely to remote teams.
  • Filtering DNS traffic to block malware, phishing, and shadow IT.
  • Enforcing device posture checks (managed devices, OS version, disk encryption).

Prerequisites

  • Cloudflare account with Zero Trust plan (free tier supports up to 50 users).
  • A domain on Cloudflare (for Access application hostnames).
  • Identity provider configured (Google Workspace, Okta, Azure AD/Entra ID, GitHub).
  • cloudflared CLI installed on the server hosting internal services.
bash
# Install cloudflared
# macOS
brew install cloudflared

# Debian/Ubuntu
curl -fsSL https://pkg.cloudflare.com/cloudflare-main.gpg | sudo tee /usr/share/keyrings/cloudflare-main.gpg >/dev/null
echo "deb [signed-by=/usr/share/keyrings/cloudflare-main.gpg] https://pkg.cloudflare.com/cloudflared $(lsb_release -cs) main" | sudo tee /etc/apt/sources.list.d/cloudflared.list
sudo apt update && sudo apt install -y cloudflared

# Docker
docker pull cloudflare/cloudflared:latest

Cloudflare Tunnel Setup

Tunnels create encrypted outbound connections from your infrastructure to Cloudflare's edge, eliminating the need to open inbound ports.

Create and Configure a Tunnel

bash
# Authenticate with Cloudflare
cloudflared tunnel login

# Create a named tunnel
cloudflared tunnel create internal-apps

# This creates credentials at ~/.cloudflared/<TUNNEL_ID>.json

# List tunnels
cloudflared tunnel list

# Route DNS to the tunnel (creates a CNAME record)
cloudflared tunnel route dns internal-apps grafana.example.com
cloudflared tunnel route dns internal-apps wiki.example.com
cloudflared tunnel route dns internal-apps ssh.example.com

Tunnel Configuration File

yaml
# ~/.cloudflared/config.yml
tunnel: <TUNNEL_ID>
credentials-file: /home/deploy/.cloudflared/<TUNNEL_ID>.json

ingress:
  # Grafana dashboard
  - hostname: grafana.example.com
    service: http://localhost:3000

  # Internal wiki
  - hostname: wiki.example.com
    service: http://localhost:8080
    originRequest:
      noTLSVerify: true

  # SSH access via browser
  - hostname: ssh.example.com
    service: ssh://localhost:22

  # Private network access (CIDR routing)
  - hostname: internal.example.com
    service: http://10.0.0.0/24

  # Catch-all — required as the last rule
  - service: http_status:404

Run the Tunnel

bash
# Run in foreground (for testing)
cloudflared tunnel run internal-apps

# Install as a systemd service
sudo cloudflared service install
sudo systemctl enable cloudflared
sudo systemctl start cloudflared

# Or run via Docker
docker run -d --name cloudflared \
  --restart unless-stopped \
  -v /home/deploy/.cloudflared:/etc/cloudflared \
  cloudflare/cloudflared:latest \
  tunnel run internal-apps

Docker Compose with Tunnel

yaml
# docker-compose.yml
version: "3.8"
services:
  cloudflared:
    image: cloudflare/cloudflared:latest
    restart: unless-stopped
    command: tunnel run
    environment:
      - TUNNEL_TOKEN=${TUNNEL_TOKEN}
    networks:
      - internal

  grafana:
    image: grafana/grafana:latest
    networks:
      - internal

  wiki:
    image: requarks/wiki:2
    networks:
      - internal

networks:
  internal:
    driver: bridge

Access Policies

Access policies control who can reach applications behind Cloudflare.

Create an Access Application

bash
# Via API — create a self-hosted application
curl -X POST "https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/access/apps" \
  -H "Authorization: Bearer $CF_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Grafana",
    "domain": "grafana.example.com",
    "type": "self_hosted",
    "session_duration": "12h",
    "auto_redirect_to_identity": true,
    "allowed_idps": ["<IDP_UUID>"]
  }'

Policy Types and Examples

bash
# Allow policy — members of the engineering group
curl -X POST "https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/access/apps/<APP_ID>/policies" \
  -H "Authorization: Bearer $CF_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Engineering Team",
    "decision": "allow",
    "include": [
      { "group": { "id": "<GROUP_UUID>" } }
    ],
    "require": [
      { "login_method": { "id": "<MFA_METHOD_UUID>" } }
    ]
  }'

Common Policy Patterns

PatternInclude RuleRequire Rule
All employeesEmail domain @company.com-
Engineering onlyAccess Group "Engineering"MFA
Contractors (time-limited)Email listDevice posture
CI/CD automationService token-
External partnersSpecific emailsCountry check

Service Tokens for Automation

bash
# Create a service token for CI/CD
curl -X POST "https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/access/service_tokens" \
  -H "Authorization: Bearer $CF_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"name": "github-actions-deploy"}'

# Response includes Client ID and Client Secret
# Use in CI with headers:
# CF-Access-Client-Id: <CLIENT_ID>
# CF-Access-Client-Secret: <CLIENT_SECRET>
bash
# Use service token in CI/CD
curl -H "CF-Access-Client-Id: $CF_CLIENT_ID" \
     -H "CF-Access-Client-Secret: $CF_CLIENT_SECRET" \
     https://grafana.example.com/api/health

Device Posture Checks

Enforce endpoint requirements before granting access.

Configure Posture Checks (Dashboard)

  1. Go to Settings > WARP Client > Device posture.
  2. Add checks:
    • Disk encryption: Require FileVault (macOS) or BitLocker (Windows).
    • OS version: Minimum macOS 14.0 or Windows 11.
    • Firewall: Ensure host firewall is enabled.
    • Crowdstrike/SentinelOne: Verify EDR agent is running.
  3. Reference posture checks in Access policies under Require rules.

Gateway DNS Filtering

Block malicious domains and enforce acceptable use policies at the DNS level.

DNS Locations

bash
# Configure DNS endpoints for offices or networks
# Dashboard: Gateway > DNS Locations > Add a location
# Assign the Gateway DNS IPs to your network's DNS resolver:
# IPv4: 172.64.36.1, 172.64.36.2
# IPv6: 2606:4700:4700::1111
# DoH: https://<UNIQUE_ID>.cloudflare-gateway.com/dns-query

DNS Policies

bash
# Create a DNS policy to block malware and phishing
curl -X POST "https://api.cloudflare.com/client/v4/accounts/$ACCOUNT_ID/gateway/rules" \
  -H "Authorization: Bearer $CF_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Block Security Threats",
    "enabled": true,
    "action": "block",
    "traffic": "any(dns.security_category[*] in {80 83 131 134 151 153})",
    "filters": ["dns"]
  }'

Common DNS Policy Rules

Rule NameTraffic ExpressionAction
Block malwareany(dns.security_category[*] in {80 83})Block
Block phishingany(dns.security_category[*] in {131 134})Block
Block social mediaany(dns.content_category[*] in {75})Block
Allow exceptionsdns.fqdn == "allowed.example.com"Allow

WARP Client Deployment

Deploy the Cloudflare WARP client to route traffic through Gateway.

bash
# MDM deployment — macOS configuration profile
# Use Cloudflare's managed deployment:
# Dashboard: Settings > WARP Client > Device enrollment

# Manual enrollment
# 1. Install WARP client from https://1.1.1.1
# 2. Click gear icon > Account > Login with Cloudflare Zero Trust
# 3. Enter your team name (from Settings > General)

# Verify WARP is connected
curl https://connectivity.cloudflare.com/cdn-cgi/trace
# Look for: warp=on

WARP Split Tunnels

bash
# Configure split tunnels to exclude certain traffic from WARP
# Dashboard: Settings > WARP Client > Device settings > Split Tunnels

# Exclude mode (default): WARP handles everything except listed IPs
# Include mode: WARP only handles listed IPs/domains

# Common exclusions:
# - Local network: 192.168.0.0/16, 10.0.0.0/8
# - Video conferencing: zoom.us, *.teams.microsoft.com
# - Printer subnets

SSH and Browser-Based Terminal

yaml
# In cloudflared config.yml — expose SSH via browser rendering
ingress:
  - hostname: ssh.example.com
    service: ssh://localhost:22
bash
# Users access ssh.example.com in their browser
# Cloudflare renders an in-browser terminal after Access authentication

# Or use cloudflared on the client side for native SSH
cloudflared access ssh --hostname ssh.example.com

# Add to SSH config for seamless access
# ~/.ssh/config
# Host ssh.example.com
#   ProxyCommand /usr/local/bin/cloudflared access ssh --hostname %h

Troubleshooting

SymptomCauseFix
Tunnel shows ERR in dashboardcloudflared not running or config errorCheck systemctl status cloudflared; validate config YAML
Access returns 403 despite correct identityPolicy order or missing require rulePolicies are evaluated top-to-bottom; ensure Allow is above Block
WARP shows "Unable to connect"Team name wrong or enrollment disabledVerify team name in Settings > General; check enrollment permissions
Service token auth failsToken expired or wrong headersRegenerate token; use both CF-Access-Client-Id and CF-Access-Client-Secret
DNS filtering not blockingClient not using Gateway DNS resolversVerify DNS is set to 172.64.36.1; check WARP is connected
Tunnel latency spikesTunnel running on overloaded hostMonitor cloudflared resource usage; run on dedicated infra
"No healthy origins" errorBackend service is downCheck the service at the configured ingress port; review cloudflared logs

Related Skills

Frequently asked questions

What does the Cloudflare Zero Trust AI skill do?

Protect internal apps with Cloudflare Access, device posture, and Zero Trust policies.

Why use Cloudflare Zero Trust on TypingMind?

Because you install it once and use it with any model. Cloudflare Zero Trust is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Cloudflare Zero Trust in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/BagelHole/DevOps-Security-Agent-Skills/tree/main/infrastructure/cloudflare/cloudflare-zero-trust. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Cloudflare Zero Trust?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Cloudflare Zero Trust?

As many as you like. As long as a model supports skills, you can use Cloudflare Zero Trust with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Cloudflare Zero Trust AI skill free?

Yes. It is published on GitHub by BagelHole under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇