Team Ownership

Prereq: read bulk-operations/SKILL.md first. JSONL piping, pagination, dry-run/digest/confirm, and hubspot history recovery live there. Reshape patterns live in bulk-operations/resources/json-patterns.md.

hubspot_owner_id is a string field on contacts, companies, deals, and tickets. Owners are CRM users — hubspot owners list returns them; there is no teams object, so team-level views are client-side groupings by hubspot_owner_id.

1. Resolve email → owner ID

Never hardcode IDs — they are portal-specific. Resolve, then cache:

FROM_ID=$(hubspot owners list | jq -r 'select(.email=="sarah@company.com") | .id')
TO_ID=$(hubspot owners list | jq -r 'select(.email=="mike@company.com")  | .id')

2. Find records for an owner

Same filter across all four object types. Add object-specific --properties for context. Unowned records use the !property form.

hubspot objects search --type contacts  --filter "hubspot_owner_id=$FROM_ID" --properties email,firstname,lifecyclestage
hubspot objects search --type companies --filter "hubspot_owner_id=$FROM_ID" --properties name,domain
hubspot objects search --type deals     --filter "hubspot_owner_id=$FROM_ID" --properties dealname,dealstage,amount
hubspot objects search --type tickets   --filter "hubspot_owner_id=$FROM_ID" --properties subject,hs_pipeline_stage

# Records with no owner at all
hubspot objects search --type deals --filter "!hubspot_owner_id" --properties dealname,amount

>100 hits — page with the --after loop from bulk-operations. Counting only: pipe to wc -l.

3. Bulk reassign — search → update

Reshape each search row into {id, properties:{hubspot_owner_id}} and pipe to objects update. Always dry-run first; for >100 rows the dry-run emits a digest + apply_command_hint — re-run with --digest/--confirm (see bulk-operations/SKILL.md § "Safe destructive workflow").

# Dry-run
hubspot objects search --type contacts --filter "hubspot_owner_id=$FROM_ID" \
| jq -c --arg to "$TO_ID" '{id, properties:{hubspot_owner_id:$to}}' \
| hubspot objects update --type contacts --dry-run

# Execute — ≤100: drop --dry-run.  >100: append --digest  --confirm .
hubspot objects search --type contacts --filter "hubspot_owner_id=$FROM_ID" \
| jq -c --arg to "$TO_ID" '{id, properties:{hubspot_owner_id:$to}}' \
| hubspot objects update --type contacts

Single-record assignment — no stdin, no jq:

hubspot objects update --type contacts 12345 --property hubspot_owner_id=$TO_ID

4. Rep-leaves workflow

Loop over the four object types the rep touches:

FROM_ID=$(hubspot owners list | jq -r 'select(.email=="leaving@company.com")    | .id')
TO_ID=$(hubspot  owners list | jq -r 'select(.email=="taking-over@company.com") | .id')

for type in contacts companies deals tickets; do
  echo "── $type ──"
  hubspot objects search --type "$type" --filter "hubspot_owner_id=$FROM_ID" \
  | jq -c --arg to "$TO_ID" '{id, properties:{hubspot_owner_id:$to}}' \
  | hubspot objects update --type "$type" --dry-run
done

Review each digest line, then re-run without --dry-run (adding --digest/--confirm per type when escalated). Mis-reassigned? hubspot history --since 1h lists the affected IDs.

5. Team-level views (client-side grouping)

Group records by hubspot_owner_id, join to owners list for human-readable emails:

hubspot objects search --type deals --filter "dealstage!=closedwon AND dealstage!=closedlost" \
  --properties hubspot_owner_id --format json \
| jq '.data | group_by(.properties.hubspot_owner_id)
       | map({owner_id: .[0].properties.hubspot_owner_id, count: length})' \
> /tmp/by-owner.json

hubspot owners list \
| jq --slurpfile by /tmp/by-owner.json -r \
     '. as $o | $by[0][] | select(.owner_id==$o.id) | "\($o.email)\t\(.count)"'

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: HubSpot
• Source: HubSpot/agent-cli-skills
• License: Apache-2.0

Install and usage instructions live in the source repository linked above.