HubSpot Services
Browse all Hubspot APIs✓ Official Vendor SpecCRMContact Managementoauth2,apiKey11 EndpointsREST
For Agents
Perform read, write, search, and batch operations on any HubSpot CRM object type — standard or custom — through a single templated /crm/v3/objects/{objectType} interface.
Use for: I need to fetch a record from a HubSpot custom object by ID, Create a record in any CRM object type without writing object-specific code, Search a custom object for records matching specific property values, Bulk-update records of a custom object during a data backfill
Not supported: Does not handle property definitions, schema creation, or marketing automation — use for record-level CRUD, batch, and search across any object type only.
The HubSpot CRM Services API provides a generic CRUD, batch, and search interface over any HubSpot CRM object type — including standard objects (companies, contacts, deals, line items, products, tickets, quotes) and custom objects defined via the Schemas API. Endpoints are templated on /crm/v3/objects/{objectType} so a single integration pattern works for every object type in the portal. The API supports single-record operations (read, update, archive), batch read, batch create, batch update, batch upsert by unique property, batch archive, and a full-featured search with filters, sorting, and paging.
Install Jentic One Beta
Connect the Services to your agent
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Services, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
1
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh2
Step 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
Capabilities
What an agent can do with Services API.
Read any CRM object record by ID via GET /crm/v3/objects/{objectType}/{objectId}
List records of any object type with pagination using GET /crm/v3/objects/{objectType}
Create a record of any object type with POST /crm/v3/objects/{objectType}
Update record properties with PATCH /crm/v3/objects/{objectType}/{objectId}
Search any object type using POST /crm/v3/objects/{objectType}/search with filters and sort
Bulk-create records with POST /crm/v3/objects/{objectType}/batch/create
Upsert records by a unique property using POST /crm/v3/objects/{objectType}/batch/upsert
Use Cases
Patterns agents use Services API for, with concrete tasks.
★ Generic CRM Object Sync Layer
Companies that integrate HubSpot with multiple internal systems prefer one templated client over many object-specific ones. The Services API exposes /crm/v3/objects/{objectType} endpoints so a single sync layer can read, write, and search standard objects (companies, deals, line items) and custom objects (loans, projects, vehicles) using the same code paths. New object types added later require zero new integration code.
Call POST /crm/v3/objects/p_loan/batch/upsert with idProperty='external_loan_id' and a list of loan records to sync from the loan origination system.
Cross-Object Data Backfill
When a new property is added to several HubSpot object types, operations needs to backfill values from a source system. Using POST /crm/v3/objects/{objectType}/batch/update against each object type, the backfill script processes records 100 at a time per object type. The same code handles companies, deals, and any custom object — only the objectType path parameter changes — keeping backfill logic compact and auditable.
For each objectType in [companies, deals, p_loan], call POST /crm/v3/objects/{objectType}/batch/update with batches of 100 records containing the new property values.
Custom Object Search Across Properties
Sales and operations teams need to filter custom object records by property combinations — e.g., all 'Project' records owned by a given user with status='In Progress' and start_date in the last 30 days. POST /crm/v3/objects/{objectType}/search supports filterGroups, sorts, and after-paging so an agent can paginate through every match. The same endpoint shape works on standard objects too, so one search routine covers the entire portal.
Call POST /crm/v3/objects/p_project/search with filterGroups for hubspot_owner_id and status, sort by start_date descending, and page through results 100 at a time.
AI Agent Working Across Custom Objects
An AI assistant agent answers questions like 'show me the top 5 active loans for Acme' regardless of whether 'loan' is standard or custom. Through Jentic the agent searches for the generic search and read operations, loads their schemas, and executes them with the right objectType. Jentic stores OAuth credentials in its vault so the agent only handles scoped tokens.
Search Jentic for 'search any hubspot crm object', load the search operation, then execute it with objectType='p_loan' and a filter on associated companyId for Acme.
Key Endpoints
11 endpoints — the hubspot crm services api provides a generic crud, batch, and search interface over any hubspot crm object type — including standard objects (companies, contacts, deals, line items, products, tickets, quotes) and custom objects defined via the schemas api.
METHOD
PATH
DESCRIPTION
GET
/crm/v3/objects/{objectType}
List records of an object type
POST
/crm/v3/objects/{objectType}
Create a record
GET
/crm/v3/objects/{objectType}/{objectId}
Read a record by ID
PATCH
/crm/v3/objects/{objectType}/{objectId}
Update a record
POST
/crm/v3/objects/{objectType}/search
Search records of an object type
POST
/crm/v3/objects/{objectType}/batch/create
Batch create records
POST
/crm/v3/objects/{objectType}/batch/upsert
Batch upsert records by unique property
GET
/crm/v3/objects/{objectType}
List records of an object type
POST
/crm/v3/objects/{objectType}
Create a record
GET
/crm/v3/objects/{objectType}/{objectId}
Read a record by ID
PATCH
/crm/v3/objects/{objectType}/{objectId}
Update a record
POST
/crm/v3/objects/{objectType}/search
Search records of an object type
POST
/crm/v3/objects/{objectType}/batch/create
Batch create records
POST
/crm/v3/objects/{objectType}/batch/upsert
Batch upsert records by unique property
Why Jentic?
Three things that make agents converge on Jentic-routed access.
Credential isolation
HubSpot OAuth tokens and private app keys are stored encrypted in the Jentic vault. Agents receive scoped execution tokens — the raw secret never enters agent context or logs.
Intent-based discovery
Agents search Jentic with intents like 'list hubspot crm records' and Jentic returns the templated Services operations with typed input schemas, so the agent can call any objectType without browsing HubSpot's docs.
Time to first call
Direct HubSpot integration with OAuth, batch handling, and per-object code paths: 3-5 days. Through Jentic: under an hour — search, load schema, execute against any object type.
Related APIs
Alternatives and complements available in the Jentic catalogue.
FAQs
Specific to using Services API through Jentic.
What authentication does the HubSpot CRM Services API use?
It supports OAuth 2.0 (recommended for public apps) and HubSpot private app access tokens passed as a Bearer token in the Authorization header. Through Jentic the token is stored encrypted in the vault and injected at execution time so the agent never holds the raw secret.
Can I use the Services API on standard objects like Contacts and Deals?
Yes. The {objectType} path parameter accepts both standard object types (contacts, companies, deals, tickets, quotes, line items, products) and custom object IDs, so the same endpoints power CRUD across all of them.
What are the rate limits for the HubSpot CRM Services API?
HubSpot enforces 100 requests per 10 seconds for OAuth apps and 190 per 10 seconds for private apps on Enterprise tiers, plus daily quotas. Use POST /crm/v3/objects/{objectType}/batch/create, batch/update, and batch/upsert (up to 100 records each) to stay efficient.
How do I upsert records by an external ID through Jentic?
Search Jentic for 'upsert hubspot crm record by external id', load POST /crm/v3/objects/{objectType}/batch/upsert, and execute it with idProperty set to your external ID property and inputs containing the records. Jentic returns typed responses for each record.
Does the search endpoint support filtering by associations?
Yes. POST /crm/v3/objects/{objectType}/search accepts filters on associated record IDs (e.g., associations.deal) alongside property filters, plus sort and paging. Use after-tokens to page through more than 200 results.
Can I list records that have been archived?
Yes. Pass archived=true on GET /crm/v3/objects/{objectType} to retrieve archived records. Without this parameter, only active records are returned.
GET STARTED