For Agents
Submit a JSON movie definition for rendering and poll its status to retrieve the resulting MP4 URL, so agents can generate personalised videos for marketing, social, or product workflows.
Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the JSON2Video 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://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | shStep 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.
What an agent can do with JSON2Video API.
Submit a structured JSON movie definition for asynchronous rendering
Poll a previously submitted movie's rendering status and result URL
Reference dynamic data values inside template scenes for personalisation
GET STARTED
Use for: I need to render a personalised video from a JSON template, Submit a movie with two scenes and a text overlay to JSON2Video, Check the status of movie ID m_abc123 to see if rendering finished, Get the MP4 URL for the latest movie I rendered
Not supported: Does not handle live video streaming, in-browser video editing, or audio-only rendering — use for asynchronous JSON-to-MP4 video generation only.
JSON2Video is a programmatic video-generation platform that compiles a structured JSON description into a rendered MP4. The API surface is intentionally small — submit a movie definition and poll for its rendering status — but the JSON template language is rich, supporting scenes, layers, transitions, text overlays, and dynamic data binding. It is the right tool when an application needs to produce social videos, personalised ads, or branded animations at scale without running ffmpeg infrastructure.
Compose multi-scene videos with transitions, text, image, and video layers
Retrieve the final MP4 URL once rendering completes for downstream delivery
Patterns agents use JSON2Video API for, with concrete tasks.
★ Personalised marketing videos at scale
Marketing teams running 1:1 video campaigns can render thousands of personalised MP4s by submitting a templated JSON movie per recipient. Each call to POST /movies returns a movie ID, and a follow-up GET /movies poll surfaces the rendered URL once compilation finishes. This avoids running an ffmpeg cluster in-house and lets the campaign tool focus on copy and CTAs.
Call POST /movies with a templated movie definition substituting the recipient's first name and product image, then poll GET /movies for the rendered URL
Programmatic social media videos
Social teams that publish daily quote, statistic, or news videos can drive the entire pipeline through JSON2Video. A scheduled job composes the JSON for the day's content, calls POST /movies, polls until complete, and uploads the resulting MP4 to TikTok, Instagram Reels, or YouTube Shorts. The structured JSON makes it easy to A/B test layouts without rewriting the rendering code.
Compose a JSON movie with the day's headline as a text layer, POST /movies, poll GET /movies, then upload the rendered MP4 to the configured social channel
Product-driven video generation
E-commerce and SaaS products that want to ship videos as a feature (e.g., 'share your year-in-review') can call JSON2Video from their backend on user request. The user-supplied data is bound into a JSON template, the resulting movie ID is stored against the user, and a webhook or polling loop surfaces the rendered URL when ready. This adds a video feature without any rendering infrastructure.
Compose a JSON movie from the user's year-in-review data, POST /movies, store the returned movieId, and surface the MP4 URL once GET /movies returns done
Agent-driven video generation via Jentic
An AI assistant asked to 'make me a 30-second product launch video' can call JSON2Video through Jentic. The agent generates a JSON movie definition, Jentic resolves the call to POST /movies, injects the API key from the vault, and returns the movie ID. A second Jentic call fetches the rendered URL for the user. This keeps the JSON2Video key out of the agent's prompt entirely.
Use Jentic to search 'render a JSON2Video movie', load POST /movies, execute with the JSON template, then poll GET /movies for the final URL
2 endpoints — json2video is a programmatic video-generation platform that compiles a structured json description into a rendered mp4.
METHOD
PATH
DESCRIPTION
/movies
Submit a movie definition for rendering
/movies
Get the status of submitted movies
/movies
Submit a movie definition for rendering
/movies
Get the status of submitted movies
Three things that make agents converge on Jentic-routed access.
Credential isolation
JSON2Video API keys are stored encrypted in the Jentic vault. Jentic injects the API key on each request at execution time, so the long-lived credential never enters the agent's prompt, conversation history, or external logs.
Intent-based discovery
Agents search Jentic with intents like 'render a video' or 'check video status' and Jentic returns the matching JSON2Video operation with its input schema, replacing manual reading of json2video.com/docs.
Time to first call
Direct JSON2Video integration: half a day to wire up auth, submission, polling, and error handling. Through Jentic: under 30 minutes — search, load schema, execute.
Alternatives and complements available in the Jentic catalogue.
Specific to using JSON2Video API through Jentic.
What authentication does the JSON2Video API use?
JSON2Video uses an API key for every request. The key is generated in the json2video.com dashboard and grants access scoped to the account. Jentic stores the key encrypted and injects it on each call so it never enters the agent's prompt context.
How long does video rendering take?
Rendering is asynchronous — POST /movies returns immediately with a movie ID, and the actual compile happens in the background. Render time depends on duration, resolution, and effects; a typical 30-second 1080p video renders within a minute or two. Poll GET /movies until status is 'done'.
What are the rate limits for the JSON2Video API?
Rate and concurrency limits are tied to the json2video.com plan tier — the spec does not publish a single numeric ceiling. Free and paid plans cap concurrent renders and monthly minutes; check the dashboard at json2video.com for current quota.
How do I render a video through Jentic?
Search Jentic for 'render a JSON2Video movie', load POST /movies, and execute with your movie definition. Jentic injects the API key. Then load GET /movies and poll until rendering completes. Sign up at https://app.jentic.com/sign-up.
Is the JSON2Video API free?
JSON2Video offers a free tier with limited monthly minutes for evaluation, plus paid plans for production volume. There is no fully free production tier; check json2video.com for current pricing and quota.
How do I retrieve the final video URL?
Call GET /movies with the movieId returned from POST /movies. When the response status is 'done', the response contains the URL to the rendered MP4. Until then the status will be 'queued' or 'rendering'.