Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Sandbox 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%2Fgetsandbox.com%2Fgetsandbox" | 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%2Fgetsandbox.com%2Fgetsandbox" | 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 Sandbox API.
Create a new mock sandbox under your account with an initial routing definition
Fork an existing sandbox to spin up an isolated copy for a feature branch
Update a sandbox's routing or stub script via PUT
Reset a sandbox's captured state to wipe assertions and recorded calls between tests
Search activity across sandboxes to inspect inbound requests by URL or time
GET STARTED
List your sandboxes and tear down unused ones with a single DELETE
Patterns agents use Sandbox API for, with concrete tasks.
★ Per-feature-branch mock servers
Engineering teams fork a baseline mock sandbox for each feature branch so frontend changes can iterate against a contract that diverges from main without touching the shared mock. The fork endpoint returns a fresh sandbox identifier the branch's CI pipeline writes into its env vars.
Fork sandbox 'payments-baseline' into 'payments-feature-x' and return the new mock URL
Reset between integration test runs
CI suites that record requests to verify behaviour need a clean slate per run. Calling DELETE /1/sandboxes/{sandboxName}/state wipes captured calls and assertions, so the next test starts from a known empty state and previous-run data does not leak across CI jobs.
Before each integration test, call DELETE /1/sandboxes/payments-mock/state to clear recorded interactions
Activity search for incident triage
When a frontend or contract test fails, engineers query sandbox activity to see exactly which inbound requests landed and what was returned. The /1/activity/search endpoint accepts URL and time filters so you can replay the failing scenario against the mock without redeploying the upstream client.
Search activity in sandbox 'payments-mock' for URLs containing '/charges' between two timestamps
AI agent for mock infrastructure ops
An ops agent connected through Jentic provisions, forks, and tears down sandboxes on demand based on chat requests like 'spin up a mock for the new auth flow'. Jentic injects the API_Key header so the agent never holds the raw credential.
Create a sandbox named 'auth-flow-mock' with a starter routing definition and return its public URL
9 endpoints — sandbox by getsandbox.
METHOD
PATH
DESCRIPTION
/1/sandboxes
List sandboxes
/1/sandboxes
Create a sandbox
/1/sandboxes/{sandboxName}
Get a sandbox
/1/sandboxes/{sandboxName}
Update a sandbox
/1/sandboxes/{sandboxName}/fork
Fork a sandbox
/1/sandboxes/{sandboxName}/state
Reset sandbox state
/1/activity/search
Search request activity
/1/sandboxes
List sandboxes
/1/sandboxes
Create a sandbox
/1/sandboxes/{sandboxName}
Get a sandbox
/1/sandboxes/{sandboxName}
Update a sandbox
/1/sandboxes/{sandboxName}/fork
Fork a sandbox
/1/sandboxes/{sandboxName}/state
Reset sandbox state
/1/activity/search
Search request activity
What agents get from Jentic-routed access to this vendor.
Setup
Wiring the Sandbox API by hand means learning its API_Key header auth, hanging your calls off the getsandbox.com/api base, and tracking each mock sandbox by name yourself. Through Jentic you install once, import the Sandbox API from the API Directory, store the API key once, and your agent calls it.
Permission scoping
Sandbox puts the sandbox name in the URL path (/1/sandboxes/{sandboxName}/...), so a rule can pin your agent to one sandbox: it can inspect and fork that sandbox and nothing else. You choose the operations it may call, so destructive ones like resetting sandbox state are not included unless you add them.
Credential isolation
Your Sandbox API key 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 'fork a mock sandbox' or 'search sandbox request activity', and Jentic returns the matching Sandbox operation with its input schema so the agent calls the right endpoint without browsing the reference docs.
Alternatives and complements available in the Jentic catalogue.
Specific to using Sandbox API through Jentic.
What authentication does the Sandbox API use?
Sandbox uses an API key sent as the API_Key header on every request. Through Jentic the key is stored encrypted in the vault and injected per request, so it never enters your agent's prompt or logs.
Can I fork a sandbox to create an isolated copy?
Yes. GET /1/sandboxes/{sandboxName}/fork returns a new sandbox cloned from the source, including its routing and stub script. This is the supported way to create per-feature-branch mocks without re-authoring the routing from scratch.
What are the rate limits for the Sandbox API?
The OpenAPI spec does not declare explicit rate limits. Practical guidance is to avoid hammering /1/activity/search in tight polling loops and to batch sandbox lifecycle operations rather than firing them in parallel from many CI jobs at once.
How do I reset a sandbox's state between test runs through Jentic?
Search Jentic for 'reset sandbox state', load the DELETE /1/sandboxes/{sandboxName}/state schema, then execute with the sandbox name. Jentic handles the API_Key injection so the reset is one structured call.
Can I search captured request activity across sandboxes?
Yes. GET /1/activity/search accepts filters for URL and time so you can locate the exact inbound request from a failing test or demo without scrolling the full request log.
Is the Sandbox API free?
Sandbox offers a free tier with limits on simultaneous sandboxes and request volume; paid plans scale up sandbox count and storage. Check getsandbox.com for current pricing before relying on the API for production-grade test infrastructure.
Can I limit what my agent is allowed to do with the Sandbox API?
Yes. Because you run Jentic One yourself, your own rules decide which Sandbox operations and credentials the agent may use. Since the sandbox name lives in the URL path (/1/sandboxes/{sandboxName}/...), you can pin the agent to a single sandbox and grant only the calls you want, such as getting and forking that sandbox while withholding others. Destructive operations like DELETE /1/sandboxes/{sandboxName}/state to reset captured state are excluded unless you explicitly add them, so the agent cannot wipe test data you did not authorize.
Know of an official OpenAPI document? Contribute it →
For Agents
Create, fork, update, and reset HTTP mock sandboxes and search activity across them through an API_Key header.
Use for: I need to create a new mock sandbox for the payments service, Fork sandbox 'payments-prod-mock' into a feature-branch sandbox, Reset the captured state of sandbox 'staging-mock' before a test run, List all sandboxes in my account
Not supported: Does not handle production traffic routing, real backend deployment, or load testing - use for managing mock sandboxes and inspecting their request activity only.
Sandbox by getsandbox.com is an HTTP mocking and stubbing service that lets developers spin up scriptable mock servers behind their own subdomain. The v1 API exposes nine endpoints covering sandbox CRUD, fork operations to clone a sandbox into a new instance, activity search across requests, and state reset for resetting captured state between test runs. It is the control surface for managing mock servers used in contract testing, demos, and frontend development against unfinished backends.