Managing DigitalOcean
Manage DigitalOcean infrastructure using the doctl CLI.
MANDATORY: Discovery-First Pattern
Always discover available resources before performing analysis.
Phase 1: Discovery
#!/bin/bash
echo "=== Account Info ==="
doctl account get --format Email,DropletLimit,Status --no-header
echo ""
echo "=== Droplets ==="
doctl compute droplet list --format ID,Name,Region,Size,VCPUs,Memory,Disk,Status,PublicIPv4 --no-header | head -30
echo ""
echo "=== Databases ==="
doctl databases list --format ID,Name,Engine,Version,Region,Status,Size,NumNodes --no-header | head -20
echo ""
echo "=== Load Balancers ==="
doctl compute load-balancer list --format ID,Name,Region,Status,IP --no-header | head -10
echo ""
echo "=== Volumes ==="
doctl compute volume list --format ID,Name,Region,Size,DropletIDs --no-header | head -20
echo ""
echo "=== Firewalls ==="
doctl compute firewall list --format ID,Name,Status --no-header | head -10
Phase 2: Analysis
#!/bin/bash
echo "=== Droplet Details ==="
DROPLET_ID="${1:?Droplet ID required}"
doctl compute droplet get "$DROPLET_ID" --format ID,Name,Region,Size,VCPUs,Memory,Disk,Status,PublicIPv4,Created
echo ""
echo "=== Droplet Bandwidth (Recent) ==="
doctl monitoring droplet bandwidth get "$DROPLET_ID" --start "$(date -u -v-1H +%Y-%m-%dT%H:%M:%SZ 2>/dev/null || date -u -d '1 hour ago' +%Y-%m-%dT%H:%M:%SZ)" --end "$(date -u +%Y-%m-%dT%H:%M:%SZ)" 2>/dev/null || echo "Bandwidth metrics not available"
echo ""
echo "=== Droplet CPU (Recent) ==="
doctl monitoring droplet cpu get "$DROPLET_ID" --start "$(date -u -v-1H +%Y-%m-%dT%H:%M:%SZ 2>/dev/null || date -u -d '1 hour ago' +%Y-%m-%dT%H:%M:%SZ)" --end "$(date -u +%Y-%m-%dT%H:%M:%SZ)" 2>/dev/null || echo "CPU metrics not available"
echo ""
echo "=== Balance & Billing ==="
doctl balance get --format MonthToDateBalance,AccountBalance,MonthToDateUsage --no-header
Output Format
RESOURCE_TYPE ID NAME REGION STATUS
droplet 12345678 web-01 nyc1 active
droplet 12345679 web-02 nyc1 active
database abc-123 pg-main nyc1 online
Safety Rules
- Use read-only commands:
list, get
- Never run
delete, destroy, remove without explicit user confirmation
- Always use
--no-header and --format for clean output
- Limit output with
| head -N to stay under 50 lines
Anti-Hallucination Rules
- NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
- NEVER fabricate metric names or dimensions — verify against the service documentation or
--help output.
- NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
- ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
- ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.
Counter-Rationalizations
| Shortcut |
Counter |
Why |
| "I'll skip discovery and check known resources" |
Always run Phase 1 discovery first |
Resource names change, new resources appear — assumed names cause errors |
| "The user only asked for a quick check" |
Follow the full discovery → analysis flow |
Quick checks miss critical issues; structured analysis catches silent failures |
| "Default configuration is probably fine" |
Audit configuration explicitly |
Defaults often leave logging, security, and optimization features disabled |
| "Metrics aren't needed for this" |
Always check relevant metrics when available |
API/CLI responses show current state; metrics reveal trends and intermittent issues |
| "I don't have access to that" |
Try the command and report the actual error |
Assumed permission failures prevent useful investigation; actual errors are informative |
1---2name: managing-digitalocean3description: Use when working with Digitalocean — digitalOcean infrastructure management via the doctl CLI. Covers Droplets, databases, load balancers, firewalls, domains, and account billing. Use when managing DigitalOcean resources, checking Droplet health, or reviewing infrastructure costs.4---56# Managing DigitalOcean78Manage DigitalOcean infrastructure using the `doctl` CLI.910## MANDATORY: Discovery-First Pattern1112**Always discover available resources before performing analysis.**1314### Phase 1: Discovery1516```bash17#!/bin/bash1819echo "=== Account Info ==="20doctl account get --format Email,DropletLimit,Status --no-header2122echo ""23echo "=== Droplets ==="24doctl compute droplet list --format ID,Name,Region,Size,VCPUs,Memory,Disk,Status,PublicIPv4 --no-header | head -302526echo ""27echo "=== Databases ==="28doctl databases list --format ID,Name,Engine,Version,Region,Status,Size,NumNodes --no-header | head -202930echo ""31echo "=== Load Balancers ==="32doctl compute load-balancer list --format ID,Name,Region,Status,IP --no-header | head -103334echo ""35echo "=== Volumes ==="36doctl compute volume list --format ID,Name,Region,Size,DropletIDs --no-header | head -203738echo ""39echo "=== Firewalls ==="40doctl compute firewall list --format ID,Name,Status --no-header | head -1041```4243### Phase 2: Analysis4445```bash46#!/bin/bash4748echo "=== Droplet Details ==="49DROPLET_ID="${1:?Droplet ID required}"50doctl compute droplet get "$DROPLET_ID" --format ID,Name,Region,Size,VCPUs,Memory,Disk,Status,PublicIPv4,Created5152echo ""53echo "=== Droplet Bandwidth (Recent) ==="54doctl monitoring droplet bandwidth get "$DROPLET_ID" --start "$(date -u -v-1H +%Y-%m-%dT%H:%M:%SZ 2>/dev/null || date -u -d '1 hour ago' +%Y-%m-%dT%H:%M:%SZ)" --end "$(date -u +%Y-%m-%dT%H:%M:%SZ)" 2>/dev/null || echo "Bandwidth metrics not available"5556echo ""57echo "=== Droplet CPU (Recent) ==="58doctl monitoring droplet cpu get "$DROPLET_ID" --start "$(date -u -v-1H +%Y-%m-%dT%H:%M:%SZ 2>/dev/null || date -u -d '1 hour ago' +%Y-%m-%dT%H:%M:%SZ)" --end "$(date -u +%Y-%m-%dT%H:%M:%SZ)" 2>/dev/null || echo "CPU metrics not available"5960echo ""61echo "=== Balance & Billing ==="62doctl balance get --format MonthToDateBalance,AccountBalance,MonthToDateUsage --no-header63```6465## Output Format6667```68RESOURCE_TYPE ID NAME REGION STATUS69droplet 12345678 web-01 nyc1 active70droplet 12345679 web-02 nyc1 active71database abc-123 pg-main nyc1 online72```7374## Safety Rules75- Use read-only commands: `list`, `get`76- Never run `delete`, `destroy`, `remove` without explicit user confirmation77- Always use `--no-header` and `--format` for clean output78- Limit output with `| head -N` to stay under 50 lines7980## Anti-Hallucination Rules81821. **NEVER assume resource names** — always discover via CLI/API in Phase 1 before referencing in Phase 2.832. **NEVER fabricate metric names or dimensions** — verify against the service documentation or `--help` output.843. **NEVER mix CLI commands between service versions** — confirm which version/API you are targeting.854. **ALWAYS use the discovery → verify → analyze chain** — every resource referenced must have been discovered first.865. **ALWAYS handle empty results gracefully** — an empty response is valid data, not an error to retry.8788## Counter-Rationalizations8990| Shortcut | Counter | Why |91|----------|---------|-----|92| "I'll skip discovery and check known resources" | Always run Phase 1 discovery first | Resource names change, new resources appear — assumed names cause errors |93| "The user only asked for a quick check" | Follow the full discovery → analysis flow | Quick checks miss critical issues; structured analysis catches silent failures |94| "Default configuration is probably fine" | Audit configuration explicitly | Defaults often leave logging, security, and optimization features disabled |95| "Metrics aren't needed for this" | Always check relevant metrics when available | API/CLI responses show current state; metrics reveal trends and intermittent issues |96| "I don't have access to that" | Try the command and report the actual error | Assumed permission failures prevent useful investigation; actual errors are informative |97
cloudthinker-ai