Managing dbt Cloud

Manage and monitor dbt Cloud — projects, environments, jobs, run history, and model artifacts.

Discovery Phase

#!/bin/bash

DBT_API="https://cloud.getdbt.com/api/v2"
AUTH="Authorization: Token $DBT_CLOUD_API_TOKEN"

echo "=== Account Info ==="
curl -s -H "$AUTH" "$DBT_API/accounts/$DBT_CLOUD_ACCOUNT_ID/" \
  | jq '.data | {name: .name, id: .id, plan: .plan, run_slots: .run_slots}'

echo ""
echo "=== Projects ==="
curl -s -H "$AUTH" "$DBT_API/accounts/$DBT_CLOUD_ACCOUNT_ID/projects/" \
  | jq -r '.data[] | [.id, .name, .connection.type, .repository.remote_url] | @tsv' | column -t | head -10

echo ""
echo "=== Environments ==="
curl -s -H "$AUTH" "$DBT_API/accounts/$DBT_CLOUD_ACCOUNT_ID/environments/" \
  | jq -r '.data[] | [.id, .name, .type, .dbt_version, .project_id] | @tsv' | column -t | head -10

echo ""
echo "=== Jobs ==="
curl -s -H "$AUTH" "$DBT_API/accounts/$DBT_CLOUD_ACCOUNT_ID/jobs/" \
  | jq -r '.data[] | [.id, .name, .project_id, .state, .schedule.cron] | @tsv' | column -t | head -15

Analysis Phase

#!/bin/bash

DBT_API="https://cloud.getdbt.com/api/v2"
AUTH="Authorization: Token $DBT_CLOUD_API_TOKEN"
ACCT="accounts/$DBT_CLOUD_ACCOUNT_ID"

echo "=== Recent Runs ==="
curl -s -H "$AUTH" "$DBT_API/$ACCT/runs/?limit=15&order_by=-created_at" \
  | jq -r '.data[] | [.id, .job.name, .status_humanized, .duration_humanized, .created_at] | @tsv' | column -t

echo ""
echo "=== Failed Runs ==="
curl -s -H "$AUTH" "$DBT_API/$ACCT/runs/?limit=10&status=20" \
  | jq -r '.data[] | [.id, .job.name, .created_at, .status_message[:60]] | @tsv' | column -t

echo ""
echo "=== Run Artifacts (latest) ==="
LATEST_RUN=$(curl -s -H "$AUTH" "$DBT_API/$ACCT/runs/?limit=1&status=10&order_by=-finished_at" | jq -r '.data[0].id')
if [ "$LATEST_RUN" != "null" ]; then
  echo "Run ID: $LATEST_RUN"
  curl -s -H "$AUTH" "$DBT_API/$ACCT/runs/$LATEST_RUN/artifacts/" \
    | jq -r '.data[]' | head -10

  echo ""
  echo "=== Run Results Summary ==="
  curl -s -H "$AUTH" "$DBT_API/$ACCT/runs/$LATEST_RUN/artifacts/run_results.json" \
    | jq '{elapsed_time: .elapsed_time, results: [.results[] | {model: .unique_id, status: .status, execution_time: .execution_time}] | sort_by(-.execution_time)[:10]}'
fi

echo ""
echo "=== Model Freshness ==="
curl -s -H "$AUTH" "$DBT_API/$ACCT/runs/$LATEST_RUN/artifacts/sources.json" 2>/dev/null \
  | jq -r '.results[] | [.unique_id, .status, .max_loaded_at, .criteria.warn_after.count // "N/A"] | @tsv' | column -t | head -10

Output Format

ACCOUNT
Name:       <account-name>
Plan:       <plan>
Run Slots:  <n>

JOBS
ID       Name             Project  State    Cron
<id>     <job-name>       <proj>   active   <cron>

RECENT RUNS
ID       Job              Status      Duration     Created
<id>     <job-name>       Success     <duration>   <timestamp>

FAILED RUNS
ID       Job              Created         Error
<id>     <job-name>       <timestamp>     <message>

MODEL PERFORMANCE (latest run)
Model              Status    Execution Time
<model-name>       pass      <seconds>s

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