How to install ticket-resolution
npx skills add https://github.com/hubspot/agent-cli-skills --skill ticket-resolutionFull instructions (SKILL.md)
Source of truth, from hubspot/agent-cli-skills.
name: ticket-resolution description: Create, triage, advance, and close HubSpot support tickets — pipeline discovery, contact/company association, priority queues, bulk stage moves, resolution close-out. triggers:
- "create ticket"
- "support ticket"
- "triage tickets"
- "open tickets"
- "resolve ticket"
- "close ticket"
- "advance ticket"
- "ticket pipeline"
- "unassigned tickets"
- "reassign ticket"
Read bulk-operations/SKILL.md first — JSONL piping, batch read, pagination, dry-run/digest/confirm, and hubspot history recovery live there. hubspot <command> --help is authoritative. Tickets use the tickets object type (plural, e.g. tickets:45123).
1. Discover pipeline + stages (portal-specific, run every session)
Stage IDs differ in every portal — never hard-code them.
hubspot pipelines list --type tickets --format table
hubspot pipelines stages --type tickets --pipeline <pipeline_id> --format table
The stage table prints each stage's ID and Label ("New", "Waiting on contact", "Closed", etc.).
2. Verify enum option values for THIS portal
hs_ticket_priority, hs_ticket_category, and hs_resolution are all enumeration properties — option values are portal-configurable and hubspot properties get does NOT return them. Discover by probing or by reading live records:
# Probe: send an invalid value; the 400 error lists the allowed options.
hubspot objects update --type tickets <some_ticket_id> --property hs_resolution=__probe__
# error: "was not one of the allowed options: [ISSUE_FIXED, FEATURE_REQUEST_TRACKED, ...]"
# Or read values already in use:
hubspot objects list --type tickets --limit 10 \
--properties hs_ticket_priority,hs_ticket_category,hs_resolution
Do NOT assume HubSpot defaults — read the portal.
3. Create a ticket and associate it
subject is the only practically-required property. Skipping hs_pipeline/hs_pipeline_stage lands the ticket in the default pipeline's first stage.
hubspot objects create --type tickets \
--property subject="Login error on mobile app" \
--property content="User reports 401 since v3.2 release." \
--property hs_pipeline=<pipeline_id> \
--property hs_pipeline_stage=<new_stage_id> \
--property hs_ticket_priority=<value_from_step_2> \
--property hs_ticket_category=<value_from_step_2>
# Capture the "id" from the output JSON.
hubspot associations create --from tickets:<ticket_id> --to contacts:<contact_id>
hubspot associations create --from tickets:<ticket_id> --to companies:<company_id>
Bulk intake from a JSONL queue (see bulk-operations/resources/json-patterns.md for reshape patterns):
cat support_requests.jsonl \
| jq -c '{properties:{subject:.subject, content:.description,
hs_pipeline:"<pipeline_id>", hs_pipeline_stage:"<new_stage_id>",
hs_ticket_priority:"<priority>", hs_ticket_category:"<category>"}}' \
| hubspot objects create --type tickets
4. Triage queries
# Open tickets by priority
hubspot objects search --type tickets \
--filter "hs_pipeline_stage=<open_stage_id> AND hs_ticket_priority=HIGH" \
--properties subject,hubspot_owner_id,createdate
# Unassigned
hubspot objects search --type tickets \
--filter "!hubspot_owner_id AND hs_pipeline_stage=<open_stage_id>" \
--properties subject,hs_ticket_priority,createdate
Filter by owner with hubspot_owner_id=<id> (find IDs via hubspot owners list --format table).
5. Advance tickets through stages (bulk update from search)
hubspot objects search --type tickets \
--filter "hs_ticket_category=BILLING_ISSUE AND hs_pipeline_stage=<new_stage_id>" \
| jq -c '{id, properties:{hs_pipeline_stage:"<waiting_stage_id>"}}' \
| hubspot objects update --type tickets --dry-run
Re-pipe the same search without --dry-run to execute. For >100 rows, follow the --digest/--confirm flow in bulk-operations/SKILL.md ("Safe destructive workflow"). Reassign in bulk works identically with {hubspot_owner_id:"<new>"}.
6. Log a resolution note
Activity creation lives in sales-execution/SKILL.md (notes/calls/meetings/tasks). After creating the note there, link it: hubspot associations create --from notes:<note_id> --to tickets:<ticket_id>.
7. Close the ticket
hs_resolution is an enumeration — pass an allowed option value from Step 2, not free text. HubSpot then computes hs_is_closed=true, closed_date, and time_to_close.
hubspot objects update --type tickets <ticket_id> \
--property hs_pipeline_stage=<closed_stage_id> \
--property hs_resolution=<allowed_resolution_value>
Known limitations
properties get/listdo not return enum options — probe via update error or read live records (CLI ask logged).- No Conversations/Inbox API surface — chat threads and inbox emails are not CLI-accessible.
Related skills
More from hubspot/agent-cli-skills and the wider catalog.

hf-cli
Hugging Face Hub CLI for managing models, datasets, spaces, buckets, and jobs on the Hub.

huggingface-best
>

huggingface-community-evals
Run evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware. Use for backend selection, local GPU evals, and choosing between vLLM / Transformers / accelerate. Not for HF Jobs orchestration, model-card PRs, .eval_results publication, or community-evals automation.

huggingface-datasets
Agent skill from huggingface/skills.

huggingface-gradio
Build Gradio web UIs and demos in Python. Use when creating or editing Gradio apps, components, event listeners, layouts, or chatbots.

huggingface-llm-trainer
Train or fine-tune language and vision models using TRL (Transformer Reinforcement Learning) or Unsloth with Hugging Face Jobs infrastructure. Covers SFT, DPO, GRPO and reward modeling training methods, plus GGUF conversion for local deployment. Includes guidance on the TRL Jobs package, UV scripts with PEP 723 format, dataset preparation and validation, hardware selection, cost estimation, Trackio monitoring, Hub authentication, model selection/leaderboards and model persistence. Use for tasks involving cloud GPU training, GGUF conversion, or when users mention training on Hugging Face Jobs without local GPU setup.