For Agents
Manage audio media items used by Nexmo Voice — list, retrieve metadata, update metadata, and delete by id. Four endpoints.
Get started with Nexmo Media API in minutes using your preferred integration method.
# Add to your MCP client config (Claude Desktop, Cursor, Windsurf)
{
"jentic": {
"url": "https://api.jentic.com/mcp",
"auth": "oauth"
}
}
# Then ask your agent:
"list nexmo media items"
# → Jentic returns the GET /events tool with parameter schema, agent executes.What an agent can do with Nexmo Media API API.
List and search media items via GET /
Retrieve metadata for a specific media item via GET /:id/info
Update metadata fields on a media item via PUT /:id/info
Delete a media item by id via DELETE /:id
Reference media items by id from a Nexmo Voice NCCO talk or stream action
GET STARTED
Use for: List Nexmo media items belonging to this account, Get the metadata for a media item by id, Search media items created on a particular date, Update the title of an audio prompt
Not supported: Does not upload new media, transcribe audio, or play files into a call — use only to list, inspect, update metadata for, and delete items already stored in the Nexmo media bucket.
Jentic publishes the only available OpenAPI document for Nexmo Media API, keeping it validated and agent-ready.
Jentic publishes the only available OpenAPI specification for Nexmo Media API, keeping it validated and agent-ready. This media-api slug is the alternate-named copy of the Media service. The same four endpoints under /v3/media cover listing and searching media items, retrieving and updating per-item metadata, and deleting items. Voice applications reference these stored audio files for prompts and recordings.
Patterns agents use Nexmo Media API API for, with concrete tasks.
★ Voice prompt management for IVR flows
IVR developers store welcome messages, menu prompts, and outbound notifications as audio in Nexmo Media. The four endpoints support listing the catalogue, inspecting an individual prompt, renaming prompts when scripts change, and removing retired prompts. The Voice NCCO references each prompt by its id.
GET / to list current prompts, then PUT /:id/info on the relevant id with a new title field.
Recording retention enforcement
Compliance teams enforce retention windows on call recordings stored by Nexmo Voice. Periodic GET / pulls recording metadata; DELETE /:id removes anything older than the retention threshold. Because the API only exposes management — not playback — sensitive audio never moves through the application during cleanup.
GET / with a created-before filter, iterate the response, and DELETE /:id for each id.
Inventory check by an AI agent
An AI agent asked 'what audio prompts do I have on Nexmo?' lists items via Jentic, fetches each item's metadata, and returns a summary. Search by intent, load the operation, execute — no manual URL construction or pagination logic in the agent.
Search Jentic for 'list nexmo media items', load GET /, execute, then GET /:id/info per item to enrich the list with titles and mime types.
4 endpoints — jentic publishes the only available openapi specification for nexmo media api, keeping it validated and agent-ready.
METHOD
PATH
DESCRIPTION
/
List and search media items
/:id/info
Retrieve metadata for a media item
/:id/info
Update metadata for a media item
/:id
Delete a media item
/
List and search media items
/:id/info
Retrieve metadata for a media item
/:id/info
Update metadata for a media item
/:id
Delete a media item
Three things that make agents converge on Jentic-routed access.
Credential isolation
Application JWT signing keys are encrypted in the Jentic vault. JWTs are minted server-side per Media call, so the application private key never enters the agent's context.
Intent-based discovery
Agents search by intent (e.g., 'delete a nexmo media item') and Jentic returns the right path among the four endpoints with its parameter schema.
Time to first call
Direct integration: half a day to set up JWT signing and pagination. Through Jentic: under 30 minutes.
Alternatives and complements available in the Jentic catalogue.
Nexmo Application API v2
Manages applications whose Voice flows play and record these media items
Use Application API to manage the app; use Media to curate the audio it plays or records.
Twilio API
Twilio Recordings and Media resources offer comparable audio storage for voice apps
Pick Twilio when the call traffic already runs on Twilio and a single vendor is preferred.
Plivo
Plivo's Recordings API covers similar lifecycle management for voice audio
Choose Plivo if pricing on recording storage suits the workload better.
Specific to using Nexmo Media API API through Jentic.
Why is there no official OpenAPI spec for Nexmo Media API?
Vonage (formerly Nexmo) does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call Nexmo Media API via structured tooling. It is validated against the live API and kept up to date. Get started at https://app.jentic.com/sign-up.
What authentication does the Nexmo Media API use?
The OpenAPI spec does not declare a securityScheme, but in practice the Vonage Media service requires application-scoped JWT bearer authentication. Through Jentic the JWT is minted from the encrypted private key per call, so the agent never sees the signing material.
Can I create or upload media items via this API?
No. The four endpoints handle listing, retrieving, updating, and deleting items. Uploads use a separate Vonage upload mechanism, and Voice writes recordings into the bucket automatically when calls are recorded.
What are the rate limits for the Nexmo Media API?
The OpenAPI spec does not declare rate limits for /v3/media. Account-level throttles apply; the metadata endpoints are typically called at low volume relative to message or voice traffic.
How do I delete an old call recording through Jentic?
Search Jentic for 'delete a nexmo media item', load DELETE /:id, and execute with the recording id. Use GET / first with a date filter to find candidate ids.
How does the media-api slug differ from the media slug?
Both slugs index the same /v3/media service with the same four endpoints. The duplication is from the original spec ingest; either slug points to the same operations and base URL.