balena Management Skill

Monitor and manage balena IoT fleet deployments and edge devices.

MANDATORY: Discovery-First Pattern

Always discover fleets and devices before querying service or release data.

Phase 1: Discovery

#!/bin/bash
BALENA_API="https://api.balena-cloud.com/v6"
AUTH="Authorization: Bearer ${BALENA_API_TOKEN}"

echo "=== User Info ==="
curl -s -H "$AUTH" "$BALENA_API/user/v1/whoami" | \
  jq -r '"Username: \(.username)\nEmail: \(.email)"' 2>/dev/null || \
curl -s -H "$AUTH" "https://api.balena-cloud.com/user/v1/whoami" | \
  jq -r '"Username: \(.username)"'

echo ""
echo "=== Fleets (Applications) ==="
curl -s -H "$AUTH" "$BALENA_API/application?\$select=app_name,slug,device_type,is_archived&\$filter=is_archived%20eq%20false" | \
  jq -r '.d[] | "\(.app_name) | Slug: \(.slug) | Device Type: \(.device_type) | Archived: \(.is_archived)"'

echo ""
echo "=== Devices ==="
curl -s -H "$AUTH" "$BALENA_API/device?\$select=device_name,uuid,is_online,os_version,supervisor_version,status&\$top=20" | \
  jq -r '.d[] | "\(.device_name) | Online: \(.is_online) | OS: \(.os_version) | Supervisor: \(.supervisor_version) | Status: \(.status)"'

echo ""
echo "=== Device Summary ==="
curl -s -H "$AUTH" "$BALENA_API/device?\$select=is_online" | \
  jq -r '"Total: \(.d | length)\nOnline: \([.d[] | select(.is_online==true)] | length)\nOffline: \([.d[] | select(.is_online==false)] | length)"'

Phase 1 outputs: User info, fleets, devices, connectivity

Phase 2: Analysis

#!/bin/bash
echo "=== Releases (latest per fleet) ==="
curl -s -H "$AUTH" "$BALENA_API/release?\$select=commit,status,created_at,release_version&\$orderby=created_at%20desc&\$top=10" | \
  jq -r '.d[] | "\(.commit[:8]) | Version: \(.release_version // "N/A") | Status: \(.status) | Created: \(.created_at)"'

echo ""
echo "=== Device OS Versions ==="
curl -s -H "$AUTH" "$BALENA_API/device?\$select=os_version" | \
  jq -r '[.d[].os_version] | group_by(.) | map({version: .[0], count: length}) | sort_by(-.count) | .[] | "\(.version): \(.count) devices"'

echo ""
echo "=== Services per Fleet ==="
curl -s -H "$AUTH" "$BALENA_API/service?\$select=service_name&\$expand=application(\$select=app_name)" | \
  jq -r '.d[] | "\(.application[0].app_name // "N/A") -> \(.service_name)"'

echo ""
echo "=== Device Environment Variables ==="
curl -s -H "$AUTH" "$BALENA_API/device_environment_variable?\$select=name,value&\$top=10" | \
  jq -r '.d[] | "\(.name)=\(.value[:30])"'

echo ""
echo "=== Update Status ==="
curl -s -H "$AUTH" "$BALENA_API/device?\$select=device_name,status,download_progress&\$filter=status%20ne%20%27Idle%27" | \
  jq -r '.d[] | "\(.device_name) | Status: \(.status) | Progress: \(.download_progress // "N/A")%"'

Output Format

BALENA STATUS
=============
Fleets: {count}
Devices: {total} (Online: {online}, Offline: {offline})
Latest Release: {commit} ({status})
OS Versions: {unique_count} in fleet
Services: {count}
Updating Devices: {count}
Issues: {list_of_warnings}

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. 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

Common Pitfalls

• OData query syntax: balena API uses OData filters — URL-encode operators
• Supervisor vs OS: Supervisor manages containers; OS is the host — update separately
• Multi-container: Each fleet can have multiple services — check all containers
• Offline devices: Devices need to be online to receive updates — track connectivity