Install Jentic One Beta

Connect the OpenTrials API to your agent

Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the OpenTrials API, 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.

Step 1: Jentic One Host machine

# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fopentrials.local%2Fopentrials" | sh

Step 2: Agent machine

# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fopentrials.local%2Fopentrials" | 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 OpenTrials API.

Run full-text search across trials, documents, and FDA applications via /search

Autocomplete trial-related entities with /search/autocomplete/{in}

Search FDA application documents via /search/fda_documents

Retrieve a specific trial by ID with /trials/{id} and inspect its source records

Look up conditions, interventions, organisations, persons, and publications by ID

GET STARTED

Start building with OpenTrials API

Explore with Jentic One

View OpenAPI Document

Use Cases

Patterns agents use OpenTrials API for, with concrete tasks.

★ Clinical Trial Discovery

Researchers and journalists investigating a drug or condition can run /search with filters to surface every registered trial and supporting document available in OpenTrials. Each trial returned exposes its source registries, intervention list, and condition list, along with /trials/{id}/records linking back to the underlying registry entries. This is useful for systematic review preparation and for spotting trial registration mismatches across registries.

GET /search with q='metformin diabetes' and inspect each returned trial's /trials/{id}/records to gather all registry entries

FDA Document Lookup

Regulatory analysts and pharma intelligence teams can use /search/fda_documents to find FDA review documents, then /fda_applications/{id} to fetch the full application record. /documents and /documents/{id} expose the cross-referenced supporting documents (protocols, results, statistical analysis plans) that OpenTrials has indexed alongside the application.

GET /search/fda_documents with q='ozempic' and resolve the top results to /fda_applications/{id} for full application metadata

Autocomplete in Trial Search UIs

Frontends building a trial search experience can wire /search/autocomplete/{in} to provide type-ahead suggestions for conditions, interventions, organisations, or persons. The {in} path parameter selects the entity type to autocomplete against, returning ranked suggestions for the user's partial input. This avoids implementing a separate suggestion index on the consumer side.

GET /search/autocomplete/conditions?q=alzhei and return the top 10 matching condition names

AI Agent Clinical Research Assistant

An AI agent helping a researcher answer 'find all trials of drug X for condition Y' can call OpenTrials through Jentic without writing endpoint URLs. Jentic resolves the agent's intent into the right /search or /trials/{id} call, dispatches the request, and returns structured JSON the agent can summarise. Because OpenTrials is unauthenticated, no key handling is needed.

Through Jentic, search 'find clinical trials about a drug', resolve to GET /search, and execute with q='pembrolizumab melanoma'

Key Endpoints

17 endpoints — opentrials api provides programmatic access to a database of clinical trial records, supporting documents, fda applications, conditions, interventions, organisations, persons, and publications.

METHOD

PATH

DESCRIPTION

GET

/search

Full-text search across trials and documents

GET

/search/autocomplete/{in}

Autocomplete suggestions by entity type

GET

/search/fda_documents

Search FDA application documents

GET

/trials/{id}

Look up a trial by ID

GET

/trials/{id}/records

Get source registry records for a trial

GET

/fda_applications/{id}

Look up an FDA application by ID

GET

/sources

List contributing data sources

GET

/search

Full-text search across trials and documents

GET

/search/autocomplete/{in}

Autocomplete suggestions by entity type

GET

/search/fda_documents

Search FDA application documents

GET

/trials/{id}

Look up a trial by ID

GET

/trials/{id}/records

Get source registry records for a trial

GET

/fda_applications/{id}

Look up an FDA application by ID

GET

/sources

List contributing data sources

Why Jentic?

What agents get from Jentic-routed access to this vendor.

Setup

Wiring the OpenTrials API by hand means constructing its search queries and paging through trial and FDA application records yourself. Through Jentic you install once, import the OpenTrials API from the API Directory, and your agent calls it directly.

Permission scoping

OpenTrials is read-only search and lookup with no credential, so scoping is by operation: you limit the agent to the operations it needs, such as searching trials or looking up an FDA application, and leave others out unless you want them. Every operation the agent can call is one you chose to allow.

Credential isolation

OpenTrials has no auth, so nothing secret is stored for this API. Jentic still dispatches each call through its audited execution path, and no credential ever enters the agent's prompt, logs, or context.

Intent-based discovery

Agents search Jentic by intent such as 'search clinical trials' or 'look up an FDA application', and Jentic returns the matching OpenTrials operation with its parameter schema so the agent calls the right endpoint without browsing the reference docs.

FAQs

Specific to using OpenTrials API through Jentic.

What authentication does the OpenTrials API use?

None - the spec defines no security schemes. Read-only requests go out unauthenticated. Through Jentic the calls dispatch directly with no vault entry needed for this API.

Can I search across multiple registries in one request with the OpenTrials API?

Yes - /search aggregates across all registries OpenTrials has indexed. The /sources endpoint enumerates the contributing registries so you can verify coverage. Each returned trial exposes its underlying source records via /trials/{id}/records.

What are the rate limits for the OpenTrials API?

OpenTrials does not document a fixed rate limit on the public spec. Treat it as a community resource and back off on transient errors. Heavy automated workloads should be run against a local mirror if available.

How do I look up a trial by ID through Jentic?

Search Jentic for 'look up clinical trial', load the GET /trials/{id} schema, and execute with the trial ID. Jentic returns the trial record including the linked /trials/{id}/records list of source registry entries.

Does the OpenTrials API include results data for trials?

Trial result publications and supporting documents are accessed through /publications/{id} and /documents/{id} respectively. The trial record itself links to these IDs so the consumer can pull results when present in the source registry.

What's the difference between /trials/{id} and /trials/{id}/records?

/trials/{id} returns the consolidated OpenTrials view of a trial, while /trials/{id}/records returns the underlying registry entries (e.g. ClinicalTrials.gov, EUCTR, ISRCTN) that were merged. /trials/{trialId}/records/{id} returns one specific source record for audit purposes.

Can I limit what my agent is allowed to do with the OpenTrials API?

Yes. Because you run Jentic One self-hosted, your own rules decide which OpenTrials operations your agent may call, and the OpenTrials API is read-only with no credential, so scoping is purely by operation. You can allow only the operations the agent needs, such as GET /search or GET /fda_applications/{id}, and leave out others like /trials/{id}/records or /sources. Every operation the agent can reach is one you chose to permit.

OpenTrials API

Name: OpenTrials API
Brand: OpenTrials API
Availability: InStock

Official vendor OpenAPI document · agent-readyHealthcareClinical Datanone17 EndpointsREST

For Agents

Search clinical trials, FDA applications, and supporting documents, and look up trials, conditions, interventions, organisations, and persons by ID.

Use for: Search clinical trials about diabetes drug X, Find documents related to a specific trial ID, Autocomplete a condition name in clinical trials, Look up an FDA application by ID

Not supported: Does not handle trial registration, patient enrollment, or adverse-event reporting - use for read-only OpenTrials search and lookup only.

OpenTrials API provides programmatic access to a database of clinical trial records, supporting documents, FDA applications, conditions, interventions, organisations, persons, and publications. Endpoints cover free-text and faceted search, autocomplete suggestions, FDA document search, and direct lookup of trials and their record provenance. The API is read-only with no authentication required, and is intended for transparency, research, and aggregation use cases that build on top of clinical trial registries.

Jentic One on Github View OpenAPI Document