cloudthinker-ai HubSpot Management Skill
Monitor and manage HubSpot CRM, marketing, and sales operations.
MANDATORY: Discovery-First Pattern
Always discover account info and object schemas before querying CRM data.
Phase 1: Discovery
#!/bin/bash
HS_API="https://api.hubapi.com"
AUTH="Authorization: Bearer ${HUBSPOT_ACCESS_TOKEN}"
echo "=== Account Info ==="
curl -s -H "$AUTH" "$HS_API/account-info/v3/details" | \
jq -r '"Portal ID: \(.portalId)\nCompany: \(.companyName // "N/A")\nTimezone: \(.timeZone)\nCurrency: \(.companyCurrency)"'
echo ""
echo "=== CRM Object Counts ==="
for obj in contacts companies deals tickets; do
count=$(curl -s -H "$AUTH" "$HS_API/crm/v3/objects/$obj?limit=1" | jq '.total // 0')
echo "$obj: $count"
done
echo ""
echo "=== Deal Pipelines ==="
curl -s -H "$AUTH" "$HS_API/crm/v3/pipelines/deals" | \
jq -r '.results[] | "\(.label) | ID: \(.id) | Stages: \(.stages | length)"'
echo ""
echo "=== Ticket Pipelines ==="
curl -s -H "$AUTH" "$HS_API/crm/v3/pipelines/tickets" | \
jq -r '.results[] | "\(.label) | ID: \(.id) | Stages: \(.stages | length)"'
echo ""
echo "=== Workflows ==="
curl -s -H "$AUTH" "$HS_API/automation/v4/flows" | \
jq -r '.results[:10] | .[] | "\(.name) | Enabled: \(.enabled) | Type: \(.type)"' 2>/dev/null || echo "Check workflows via dashboard"
Phase 1 outputs: Account info, object counts, pipelines, workflows
Phase 2: Analysis
#!/bin/bash
echo "=== Deal Pipeline Summary ==="
curl -s -H "$AUTH" "$HS_API/crm/v3/objects/deals?limit=100&properties=dealstage,amount,pipeline" | \
jq -r '[.results[] | {stage: .properties.dealstage, amount: (.properties.amount // "0" | tonumber)}] | group_by(.stage) | map({stage: .[0].stage, count: length, total: [.[].amount] | add}) | .[] | "\(.stage): \(.count) deals (\(.total))"'
echo ""
echo "=== Recent Contacts (last 7 days) ==="
curl -s -H "$AUTH" "$HS_API/crm/v3/objects/contacts?limit=1&sorts=-createdate&filterGroups=[{\"filters\":[{\"propertyName\":\"createdate\",\"operator\":\"GTE\",\"value\":\"$(date -v-7d +%s000 2>/dev/null || date -d '7 days ago' +%s000)\"}]}]" | \
jq -r '"New Contacts (7d): \(.total // "N/A")"' 2>/dev/null
echo ""
echo "=== Email Campaign Performance ==="
curl -s -H "$AUTH" "$HS_API/marketing/v3/emails/statistics?limit=5&orderBy=-sendDate" | \
jq -r '.results[:5] | .[] | "\(.name[:40]) | Sent: \(.counters.sent) | Opens: \(.counters.open) | Clicks: \(.counters.click)"' 2>/dev/null || echo "Check marketing email stats via dashboard"
echo ""
echo "=== Forms ==="
curl -s -H "$AUTH" "$HS_API/marketing/v3/forms" | \
jq -r '.results[:10] | .[] | "\(.name) | Type: \(.formType) | Submissions: \(.submissions // "N/A")"' 2>/dev/null
echo ""
echo "=== Integration Health ==="
curl -s -H "$AUTH" "$HS_API/integrations/v1/me" | \
jq -r '"App ID: \(.appId // "direct token")\nScopes: \(.scopes // [] | join(", "))"' 2>/dev/null
Output Format
HUBSPOT STATUS
==============
Portal: {id} ({company})
Contacts: {count} | Companies: {count}
Deals: {count} | Tickets: {count}
Deal Pipelines: {count}
Pipeline Value: {total_amount}
Active Workflows: {count}
New Contacts (7d): {count}
Issues: {list_of_warnings}
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 |
Common Pitfalls
- Rate limits: 100 requests per 10 seconds for OAuth apps — use batch endpoints
- API versions: CRM v3 is current — v1/v2 endpoints are deprecated
- Property limits: Only 10 properties returned by default — specify properties parameter
- Associations: Objects are linked via associations API — separate from object queries
- OAuth scopes: Each API endpoint requires specific scopes — check token permissions