Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the ROOTE 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%2Froote.ai%2Froote" | shStep 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%2Froote.ai%2Froote" | 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.
What an agent can do with ROOTE API.
Search mobility, transit, and services families near a lat and lon in one call with families and family_limit
Read upcoming departures for a canonical transport station, filtered by line and direction
Plan an A-to-B public transport journey with max_transfers, wheelchair, and a provider_strategy of primary, fallback, or compare
Calculate a journey through ordered waypoints with the POST journey form
GET STARTED
Find nearby EV charging stations with connector availability where supplied
Locate nearby parking and shared mobility within a radius
Geocode an address into up to five candidates or reverse-geocode coordinates
Patterns agents use ROOTE API for, with concrete tasks.
★ Travel assistant that answers how do I get there
A travel agent asked how to get somewhere needs live options, not a remembered timetable. Through Jentic it searches for 'plan a public transport journey', loads ROOTE's getTransportJourney, and passes origin_lat, origin_lon, destination_lat, and destination_lon with an optional departure_time, while the bearer token is added by the developer's own instance. It returns alternatives with transfers and scheduled versus realtime times.
Call GET /v1/transit/journey with origin_lat 52.52, origin_lon 13.405, destination_lat 52.5096, destination_lon 13.3759, departure_time '2026-11-18T09:00:00+01:00', and max_transfers 2, then summarise the fastest option
Live departures at the nearest stop
Someone standing on a street corner wants the next train or bus. GET /v1/transit/nearby returns stations around a coordinate, and the canonical station id feeds GET /v1/transit/stations/{id}/departures, which returns upcoming departures with expected_at times and can be filtered by line or direction.
Call GET /v1/transit/nearby with lat 40.7527 and lon -73.9772, take the closest station id, then call GET /v1/transit/stations/{id}/departures with limit 5
EV driver charging and parking finder
An in-car or trip assistant helps a driver find somewhere to charge or park near their destination. GET /v1/charging/nearby returns Google Places EV charging stations with connector availability when supplied, never reporting missing availability as zero, and GET /v1/parking/nearby lists parking around the same coordinate.
Call GET /v1/charging/nearby with lat 51.5072, lon -0.1276, and radius 3000, then call GET /v1/parking/nearby for the same point and return both lists
One-call mobility snapshot for a location
A city app showing everything nearby can use GET /v1/nearby, which with no filters queries mobility, transit, and services families together. The families parameter narrows the search, family_limit caps results per family, and one credit is settled for each family that returns, with a total technical failure of a family refunded.
Call GET /v1/nearby with lat 41.3874, lon 2.1686, families 'mobility,transit', and family_limit 5
20 endpoints — the roote api finds live mobility options around a coordinate anywhere in the world: shared mobility, public transport stations and departures, urban services, parking, and ev charging.
METHOD
PATH
DESCRIPTION
/v1/nearby
Search all nearby mobility families
/v1/transit/nearby
Find nearby transport stations
/v1/transit/stations/{id}/departures
Upcoming departures for a station
/v1/transit/journey
Calculate an A-to-B transit journey
/v1/transit/journey
Calculate a journey with waypoints
/v1/charging/nearby
Find nearby EV charging
/v1/parking/nearby
Find nearby parking
/v1/geocode
Geocode an address
/v1/nearby
Search all nearby mobility families
/v1/transit/nearby
Find nearby transport stations
/v1/transit/stations/{id}/departures
Upcoming departures for a station
/v1/transit/journey
Calculate an A-to-B transit journey
/v1/transit/journey
Calculate a journey with waypoints
/v1/charging/nearby
Find nearby EV charging
/v1/parking/nearby
Find nearby parking
/v1/geocode
Geocode an address
What agents get from Jentic-routed access to this vendor.
Setup
Wiring ROOTE by hand means sorting out which of its 20 operations need the rt_live_ bearer token, which accept anonymous calls, and which scopes like journey:read apply, plus handling the lon versus lng aliases across calls. Through Jentic you install once, import ROOTE from the API Directory, store the token once, and your agent calls it.
Permission scoping
ROOTE puts station and line ids in the URL path (/v1/transit/stations/{id}/departures), so a rule can pin your agent to departures for one station, and most other inputs are query parameters. You choose the operations it may call, so ones that spend more credits, like POST /v1/capabilities or ride-hailing estimates, are not included unless you add them.
Credential isolation
Your ROOTE token is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'transit stations near me' or 'plan a journey', and Jentic returns the matching ROOTE operation with its input schema, including required coordinates like origin_lat and destination_lon.
Alternatives and complements available in the Jentic catalogue.
Specific to using ROOTE API through Jentic.
What authentication does the ROOTE API use?
The ROOTE OpenAPI spec declares bearerAuth, an HTTP bearer token in the rt_live_ format. Parking, EV charging, journey, ride-hailing, and capability calls list bearer as required, while nearby, geocoding, place search, and station reads accept either no credential or the bearer, and health, changelog, and city usage routes are open. Several operations also take a token query parameter that the spec calls a compatibility API token, preferring the Authorization header, and some descriptions name scopes such as journey:read and data:read.
Can I plan a journey with stops along the way using the ROOTE API?
Yes. POST /v1/transit/journey takes origin and destination, which are required, plus ordered waypoints, constraints, modes, and alternatives in the body. For providers without native waypoint support, ROOTE calls each segment in turn and checks continuity between them.
What are the rate limits for the ROOTE API?
The OpenAPI spec does not specify request rate limits; it describes per-family credit settlement on the nearby search and plan limits on result counts. Check https://roote.ai/ for current plan limits.
Is there a ROOTE MCP server?
You do not need an MCP server to give your agent ROOTE. Jentic connects it directly from the API Directory: import it, store your credential once, and your agent finds nearby search, departures, and journey operations by intent when it needs them.
How do I get live departures through Jentic?
Search Jentic for 'next departures from a station', load getTransportDepartures (GET /v1/transit/stations/{id}/departures), and pass the canonical station id from a nearby search, with optional line, direction, and limit. To run it on your own infrastructure, install Jentic One from its GitHub repo.
Can I limit what my agent is allowed to do with the ROOTE API?
Yes. For a commuter helper, write a rule that allows GET /v1/transit/nearby and GET /v1/transit/stations/{id}/departures only. The agent can find stops and read departures but cannot run journey planning, ride-hailing estimates, or capability calls, and every call it makes is logged.
For Agents
Find transit, shared mobility, parking, and EV charging near a coordinate, read live departures, geocode addresses, and plan public transport journeys with ROOTE.
Use for: Find all transit stations near 48.8566, 2.3522, Get the next departures from this station on line 4, Plan a wheelchair-accessible transit journey from my hotel to the airport, Search for EV charging stations within 2 km
Not supported: Does not book tickets, hail or pay for rides, or render map tiles. Use for finding live mobility, departures, and journeys near a coordinate only.
The ROOTE API finds live mobility options around a coordinate anywhere in the world: shared mobility, public transport stations and departures, urban services, parking, and EV charging. A single nearby call fans out across the mobility, transit, and services families, while dedicated calls cover each one. It also geocodes and reverse-geocodes addresses, searches places, plans A-to-B public transport journeys with optional waypoints, and returns ride-hailing estimates. Station and line ids are canonical ROOTE ids, so a station found nearby can be passed straight to its departures call.
This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.
Base layer of spec validity and structural soundness.
Aggregated quality score from linter diagnostics, weighted by severity.
Percentage of `$ref` references that resolve successfully.
Checks whether the API description parses successfully and conforms to its declared specification (e.g., OpenAPI).
Structural correctness score based on schema issues using logarithmic dampening.
Clarity, completeness, and ingestion readiness for developers and tooling.
How richly the API is illustrated with examples.
Percentage of examples that conform to their schemas.
Percentage of operations with complete response definitions (success, client error, server error).
Health of API ingestion, bundling, and resolution within Jentic pipelines.
Semantic breadth, depth, and agent comprehension for AI systems.
Coverage of descriptions across API elements.
Coverage of RFC 9457 Problem Details for error responses.
Coverage, uniqueness, and casing consistency of operationIds for AI inference.
Coverage of summaries across operations/tags/info.
Functional utility, complexity comfort, and AI orchestration readiness.
Agent comfort level based on API operational and structural complexity.
Trust, risk posture, and security compliance.
Average quality of security schemes based on authentication method strength (weakest link for OAuth2).
Findability, semantic richness, and reasoning readiness.
Clarity and depth of descriptions across API elements.
Score it yourself
Every API in the directory is allowlisted, so you can re-score it with no key required.
npx @jentic/api-scorecard-cli score <openapi-url>