Oracle Cloud Infrastructure (OCI)

CRITICAL MUST: Secrets and .env

• •NEVER store live .env files or credentials inside any skill folder.
• •.env.template files belong only in templates/ within a skill.
• •Store live secrets in ~/.admin/.env (or another non-skill location you control) and reference them from there.

Status: Production Ready | Dependencies: OCI CLI, SSH key pair

Navigation

• •Operations, troubleshooting, config, and cleanup: references/OPERATIONS.md
• •CLI install/config/capacity/networking details: docs/

Step 0: Gather Required Information (MANDATORY)

STOP. Before ANY deployment commands, collect ALL parameters from the user.

Copy this checklist and confirm each item:

Required Parameters:
- [ ] SERVER_NAME       - Unique name for this server
- [ ] OCI_REGION        - Region (us-ashburn-1, us-phoenix-1, ca-toronto-1, etc.)
- [ ] OCI_SHAPE         - Instance shape (see profiles below)
- [ ] OCI_OCPUS         - Number of OCPUs (1-4 for Always Free)
- [ ] OCI_MEMORY_GB     - Memory in GB (1-24 for Always Free)
- [ ] SSH_KEY_PATH      - Path to SSH public key (default: ~/.ssh/id_rsa.pub)

Deployment Purpose (determines recommended profile):
- [ ] Purpose: coolify / kasm / both / custom
      coolify → 2 OCPU, 12GB RAM (Always Free eligible)
      kasm    → 4 OCPU, 24GB RAM (Always Free eligible)
      both    → 4 OCPU, 24GB RAM (Always Free eligible)

Always Free Tier Limits:
- Shape: VM.Standard.A1.Flex (ARM64)
- Max: 4 OCPUs, 24GB RAM total (can split across instances)

Recommended profiles by purpose:

DO NOT proceed to Prerequisites until ALL parameters are confirmed.

Prerequisites

Before using this skill, verify the following:

1. OCI CLI Installed

oci --version

If missing, install with:

bash -c "$(curl -L https://raw.githubusercontent.com/oracle/oci-cli/master/scripts/install/install.sh)" -- --accept-all-defaults
source ~/.bashrc  # or restart terminal

2. OCI CLI Configured

ls ~/.oci/config

If missing, configure with:

oci setup config

You'll need:

• •Tenancy OCID (OCI Console → Profile → Tenancy)
• •User OCID (OCI Console → Profile → My Profile)
• •Region (e.g., us-ashburn-1)
• •API key pair (wizard generates this)

3. SSH Key Pair

ls ~/.ssh/id_rsa.pub

If missing, generate with:

ssh-keygen -t rsa -b 4096 -f ~/.ssh/id_rsa -N ""

4. SSH Key Permissions

stat -c %a ~/.ssh/id_rsa  # Should be 600

If wrong, fix with:

chmod 600 ~/.ssh/id_rsa

5. Test Authentication

oci iam availability-domain list

If this fails: API key may not be uploaded to OCI Console → Profile → API Keys

Quick Start

1. Check Capacity First

OCI Always Free ARM instances are highly demanded. Always check before deploying:

./scripts/check-oci-capacity.sh

# Check specific region
./scripts/check-oci-capacity.sh us-ashburn-1

# Use different OCI profile
./scripts/check-oci-capacity.sh --profile PRODUCTION

# Check multiple regions
for region in us-ashburn-1 us-phoenix-1 ca-toronto-1; do
  echo "=== $region ==="
  ./scripts/check-oci-capacity.sh "$region"
done

No capacity? Use automated monitoring:

./scripts/monitor-and-deploy.sh --stack-id <STACK_OCID>

2. Deploy Infrastructure

# Configure environment
cp scripts/.env.example scripts/.env
# Edit scripts/.env with your values

# Deploy
./scripts/oci-infrastructure-setup.sh

Scripts Reference

check-oci-capacity.sh

Checks VM.Standard.A1.Flex availability across availability domains.

./scripts/check-oci-capacity.sh                    # Home region
./scripts/check-oci-capacity.sh us-ashburn-1       # Specific region
./scripts/check-oci-capacity.sh --profile DANIEL   # With profile

Tests 4 OCPU / 24GB RAM (full free tier), falls back to 2/12 if unavailable.

monitor-and-deploy.sh

Continuously monitors and auto-deploys when capacity found.

./scripts/monitor-and-deploy.sh \
  --stack-id <STACK_OCID> \
  --profile DANIEL \
  --interval 300 \
  --max-attempts 100

oci-infrastructure-setup.sh

Creates complete infrastructure: compartment → VCN → subnet → IGW → security list → instance.

Requires .env file with OCI credentials and configuration.

cleanup-compartment.sh

Safely deletes compartment and all resources (requires confirmation).

./scripts/cleanup-compartment.sh ocid1.compartment.oc1..xxx

Operations

Common issues, best practices, configuration variables, and cleanup ordering are in references/OPERATIONS.md.

Logging Integration

When performing infrastructure operations, log to the centralized system:

# After provisioning
log_admin "SUCCESS" "operation" "Provisioned OCI instance" "id=$INSTANCE_ID provider=OCI"

# After destroying
log_admin "SUCCESS" "operation" "Terminated OCI instance" "id=$INSTANCE_ID"

# On error
log_admin "ERROR" "operation" "OCI deployment failed" "error=OUT_OF_HOST_CAPACITY"

See admin skill's references/logging.md for full logging documentation.

Additional Documentation

• •Installation Guide - Install OCI CLI on any OS
• •Configuration Guide - Set up OCI credentials
• •Capacity Guide - Handling capacity issues
• •Networking Guide - VCN, subnets, security lists
• •Troubleshooting Guide - Common issues and solutions

Official Resources

• •OCI Documentation
• •OCI CLI Reference
• •Always Free Tier
• •ARM Instances Guide