OSDB REST API v1

Name: OSDB REST API v1 API
Brand: OSDB REST API v1
Availability: InStock

★ Only Publicly Available OpenAPI DocumentAI/MLLanguage Modelsnone10 EndpointsREST

For Agents

Discover, load, and execute OpenLink Smart Data Bot services and their actions to drive conversational data workflows over HTTP.

Connect the OSDB REST API v1 to your agent

Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the OSDB REST API v1, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.

Install Jentic One Beta (Self-Hosted)

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 | sh

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 instance

Jentic 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 OSDB REST API v1 API.

List the Smart Data Bot services currently available on the OSDB instance

Load a service so its actions can be invoked by clients

Inspect the action set for a loaded service, including help text and parameters

Execute a named action on a service and receive its structured result

GET STARTED

Start building with OSDB REST API v1 API

Explore with Jentic

View OpenAPI Document

Jentic publishes the only available OpenAPI document for OSDB REST API v1, keeping it validated and agent-ready.

The OpenLink Smart Data Bot (OSDB) REST API v1 lets clients enumerate, load, and unload conversational services and invoke their actions over HTTP. Each service exposes a set of named actions with help and parameter schemas, and the /api/v1/actions/{serviceId}/{actionId}/exec endpoint runs the action and returns the result. Authentication is handled by login and logout endpoints, and the API is the programmatic surface behind OpenLink's Smart Data Bot tooling.

Use Cases

Patterns agents use OSDB REST API v1 API for, with concrete tasks.

★ Programmatic Smart Data Bot Action Execution

Drive an OpenLink Smart Data Bot from a custom UI or automation by listing services, loading the one needed, then calling its actions through POST /api/v1/actions/{serviceId}/{actionId}/exec. The action's parameter schema is discoverable via the help endpoint, so clients can build dynamic forms or agent prompts without hard-coded shapes.

Load the service 'sql-agent', list its actions, and execute the action that takes a SQL query parameter against a target dataset, then return the action's response.

Service Lifecycle Management

Operators can manage which Smart Data Bot services are loaded on a host using POST and DELETE on /api/v1/services. This is useful when deploying a new service version or when conserving resources by keeping only currently used services loaded.

Unload the service with ID 'demo-bot' via DELETE /api/v1/services/{serviceId} and confirm via GET /api/v1/services that it no longer appears in the loaded list.

Action Discovery for an Agent UI

Build an agent UI that introspects available services and their actions on the fly, presenting users with the operations they can run. The agent calls GET /api/v1/services to list services, then GET /api/v1/actions/{serviceId} to enumerate actions, and uses /help to surface human-readable parameter descriptions.

List loaded services, pick the first one, list its actions, and render each action's help text so a user can choose one to run.

AI Agent Tool Use via Jentic

An LLM agent can treat OSDB actions as tools, discovering them through Jentic and calling exec endpoints with structured arguments. Jentic returns the action's input schema so the agent does not have to parse OpenAPI, and the call result is forwarded back to the agent as part of its reasoning trace.

Use Jentic to search for 'execute an osdb action', load the exec operation for the chosen serviceId and actionId, and run it with the supplied JSON arguments.

Key Endpoints

10 endpoints — the openlink smart data bot (osdb) rest api v1 lets clients enumerate, load, and unload conversational services and invoke their actions over http.

METHOD

PATH

DESCRIPTION

GET

/api/v1/services

List loaded services

POST

/api/v1/services

Load a service

DELETE

/api/v1/services/{serviceId}

Unload a service

GET

/api/v1/actions/{serviceId}

List actions for a service

POST

/api/v1/actions/{serviceId}/{actionId}/exec

Execute a named action

GET

/api/v1/actions/{serviceId}/{actionId}/help

Get help and parameter info for an action

GET

/api/v1/services

List loaded services

POST

/api/v1/services

Load a service

DELETE

/api/v1/services/{serviceId}

Unload a service

GET

/api/v1/actions/{serviceId}

List actions for a service

POST

/api/v1/actions/{serviceId}/{actionId}/exec

Execute a named action

Why though Jentic?

Three things that make agents converge on Jentic-routed access.

Credential isolation

OSDB uses session-based login and logout endpoints rather than a header-based API key. Jentic vaults the credentials used at login and can transparently replay the session token, so the agent never sees raw secrets.

Intent-based discovery

Agents search Jentic with intents like 'list osdb services' or 'execute an osdb action' and receive the matching operation with its parameter schema, removing the need to parse the spec.

Time to first call

Direct integration: 1-2 days to wire up service load, action discovery, and execution flows. Through Jentic: under 1 hour.

Related APIs

Alternatives and complements available in the Jentic catalogue.

Alternative

OSDB REST API v1 (alt)

A second packaging of the same OSDB v1 surface; identical operations under a different api_slug.

Choose this slug when integrating from a different OpenLink-managed catalogue entry; the operations are equivalent.

Complementary

Pinecone API

Pinecone provides vector storage that an OSDB action could query as a downstream step in a data workflow.

Pair the two when the OSDB action is implemented over a vector retrieval step.

Complementary

Elastic API

Elastic offers full-text and analytics search; OSDB actions can be designed to call it as a backing store.

Use Elastic for the data layer and OSDB for the action layer that orchestrates queries.

FAQs

Specific to using OSDB REST API v1 API through Jentic.

What authentication does the OSDB REST API use?

The OSDB API exposes GET /api/v1/login and GET /api/v1/logout endpoints to manage a session and the spec does not declare a securityScheme, so no API key is required to call most endpoints. Through Jentic, any session token returned by login can be vaulted and replayed automatically.

Can I execute a Smart Data Bot action over HTTP with the OSDB REST API?

Yes. POST /api/v1/actions/{serviceId}/{actionId}/exec runs the named action on the loaded service and returns its result. Use the /help endpoint on the same path to retrieve parameter documentation before constructing the request body.

What are the rate limits for the OSDB REST API?

OSDB is typically self-hosted by OpenLink customers, so rate limits depend on the host's deployment. The published spec does not declare a hard limit; treat it as fair-use against the running instance and check with the operator for production quotas.

How do I list available services with the OSDB REST API through Jentic?

Search Jentic for 'list openlink smart data bot services', load GET /api/v1/services, and execute it. Jentic returns the structured list of services along with their identifiers.

What is the difference between loading and executing a service in the OSDB REST API?

Loading a service via POST /api/v1/services makes its actions available on the host. Executing an action via POST /api/v1/actions/{serviceId}/{actionId}/exec runs the chosen action with input parameters. A service must be loaded before its actions can be executed.