Managing Hightouch
Manage and monitor Hightouch reverse ETL — syncs, destinations, models, and sync run health.
Discovery Phase
#!/bin/bash
HIGHTOUCH_API="https://api.hightouch.com/api/v1"
AUTH="Authorization: Bearer $HIGHTOUCH_API_KEY"
echo "=== Sources ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/sources" \
| jq -r '.data[] | [.id, .name, .type, .slug] | @tsv' | column -t | head -10
echo ""
echo "=== Destinations ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/destinations" \
| jq -r '.data[] | [.id, .name, .type, .slug] | @tsv' | column -t | head -10
echo ""
echo "=== Models ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/models" \
| jq -r '.data[] | [.id, .name, .slug, .sourceId, .primaryKey] | @tsv' | column -t | head -15
echo ""
echo "=== Syncs ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs" \
| jq -r '.data[] | [.id, .slug, .destinationId, .modelId, .schedule.type, .disabled] | @tsv' | column -t | head -15
Analysis Phase
#!/bin/bash
HIGHTOUCH_API="https://api.hightouch.com/api/v1"
AUTH="Authorization: Bearer $HIGHTOUCH_API_KEY"
echo "=== Recent Sync Runs ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs/runs?limit=15&orderBy=createdAt&order=desc" \
| jq -r '.data[] | [.id, .syncId, .status, .recordsProcessed, .recordsFailed, .completedAt] | @tsv' | column -t
echo ""
echo "=== Failed Sync Runs ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs/runs?status=failed&limit=10" \
| jq -r '.data[] | [.id, .syncId, .startedAt, .error[:60]] | @tsv' | column -t
echo ""
echo "=== Sync Performance Summary ==="
for SYNC_ID in $(curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs" | jq -r '.data[:5][].id'); do
SYNC_SLUG=$(curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs/$SYNC_ID" | jq -r '.data.slug')
LAST_RUN=$(curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs/$SYNC_ID/runs?limit=1" \
| jq '{status: .data[0].status, processed: .data[0].recordsProcessed, failed: .data[0].recordsFailed, duration: .data[0].duration}')
echo "$SYNC_SLUG: $LAST_RUN"
done
echo ""
echo "=== Sync Configuration ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs/$HIGHTOUCH_SYNC_ID" \
| jq '{slug: .data.slug, schedule: .data.schedule, destination: .data.destinationId, model: .data.modelId, configuration: .data.configuration | keys}'
echo ""
echo "=== Sync Alerts ==="
curl -s -H "$AUTH" "$HIGHTOUCH_API/syncs/$HIGHTOUCH_SYNC_ID/alerts" \
| jq -r '.data[] | [.type, .status, .threshold, .lastTriggeredAt] | @tsv' | column -t | head -5
Output Format
SOURCES
ID Name Type Slug
<id> <source-name> <type> <slug>
DESTINATIONS
ID Name Type Slug
<id> <dest-name> <type> <slug>
SYNCS
ID Slug Dest ID Model ID Schedule Disabled
<id> <sync-slug> <did> <mid> <type> false
SYNC RUNS
ID Sync ID Status Processed Failed Completed
<id> <sid> success <n> <n> <timestamp>
FAILED RUNS
ID Sync ID Started Error
<id> <sid> <timestamp> <message>
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
--helpoutput. - 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 |