For Agents
Generate complete PowerPoint presentations from raw text, a topic, a summary or a YouTube URL through MagicSlides' four AI generation endpoints.
Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the MagicSlides 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 MagicSlides API.
Generate a full presentation from arbitrary input text via POST /ppt-from-text
Produce a slide deck from a single topic line through the legacy /ppt_from_topic endpoint
Convert a long-form summary into a deck via the legacy /ppt_from_summery endpoint
GET STARTED
Use for: I need to generate a slide deck from a meeting summary, Turn a YouTube tutorial into a presentation for our team, Create a 10-slide deck from a single topic line, Generate a Spanish-language presentation from English source text
Not supported: Does not handle slide editing, real-time collaboration, or hosting of generated decks — use for one-shot AI generation of presentation files only.
Jentic publishes the only available OpenAPI specification for MagicSlides API, keeping it validated and agent-ready. MagicSlides is an AI presentation generator that turns text, summaries, topics or YouTube videos into professional slide decks in seconds and supports more than 100 languages. The public API exposes four POST endpoints, with /ppt-from-text as the current canonical generator and three deprecated variants kept for backward compatibility.
Generate a slide deck from a YouTube video transcript through the legacy /ppt_from_youtube endpoint
Produce decks in over 100 languages by setting the language field on the input payload
Patterns agents use MagicSlides API for, with concrete tasks.
★ Meeting Notes to Slide Deck
Convert long-form meeting notes or a written briefing into a presentation ready to share. POST /ppt-from-text accepts the source text, the desired number of slides and a target language, and returns the generated deck. Suitable for chiefs of staff, founders and product managers who need to brief stakeholders without spending an hour authoring slides manually.
POST /ppt-from-text with the meeting transcript and slide_count=10 to receive the rendered presentation
YouTube Video to Presentation
Repackage a YouTube tutorial as an internal training deck. The legacy POST /ppt_from_youtube endpoint accepts a video URL and produces a slide summary suitable for handouts. Useful for L&D teams curating external content into structured internal courses without re-recording or transcribing manually.
POST /ppt_from_youtube with the training video's URL to receive a slide deck for the L&D library
Sales Pitch Generator
Spin up a tailored sales pitch on demand from a topic line such as 'cybersecurity for healthcare CIOs'. POST /ppt_from_topic returns a topical deck without the seller needing to script every slide manually. Suitable for SDR teams and outbound campaigns where the cost of a custom deck per prospect would otherwise be prohibitive.
POST /ppt_from_topic with topic 'cybersecurity for healthcare CIOs' and language='en' to receive a tailored pitch deck
AI Agent Slide Generation
Let an AI assistant produce a polished deck on demand inside a chat or workflow without holding the MagicSlides API key. Through Jentic the agent searches for the slide generation operation, loads the schema for /ppt-from-text, and executes with credentials brokered by the vault. Useful for executive assistants, content production agents and BI assistants that need a presentation as the final artefact.
Search Jentic for 'generate a presentation from text', load POST /ppt-from-text, and execute with the user's source notes and the requested slide count
4 endpoints — jentic publishes the only available openapi specification for magicslides api, keeping it validated and agent-ready.
METHOD
PATH
DESCRIPTION
/ppt-from-text
Generate a presentation from input text
/ppt_from_topic
Generate a presentation from a single topic (deprecated)
/ppt_from_summery
Generate a presentation from a summary (deprecated)
/ppt_from_youtube
Generate a presentation from a YouTube URL (deprecated)
/ppt-from-text
Generate a presentation from input text
/ppt_from_topic
Generate a presentation from a single topic (deprecated)
/ppt_from_summery
Generate a presentation from a summary (deprecated)
/ppt_from_youtube
Generate a presentation from a YouTube URL (deprecated)
Three things that make agents converge on Jentic-routed access.
Credential isolation
MagicSlides API keys are stored encrypted in the Jentic vault (MAXsystem). Agents receive scoped access tokens — the raw key never enters the agent's context, so prompt leakage cannot expose the underlying credential.
Intent-based discovery
Agents search by intent (e.g. 'generate a presentation from text' or 'convert a youtube video to slides') and Jentic returns the matching MagicSlides operation along with its input schema, so the agent picks /ppt-from-text or one of the legacy endpoints without reading docs.
Time to first call
Direct MagicSlides integration: 2-3 hours for auth, payload shaping and async download handling. Through Jentic: under 30 minutes — search, load schema, execute.
Alternatives and complements available in the Jentic catalogue.
Specific to using MagicSlides API through Jentic.
Why is there no official OpenAPI spec for MagicSlides API?
MagicSlides publishes a documentation page at magicslides.app/magicslides-api/docs but does not ship a curated OpenAPI specification. Jentic generates and maintains this spec so AI agents and developers can call MagicSlides 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 MagicSlides API use?
The API uses an API key passed in a request header (the apiKey scheme defined in the spec). Through Jentic the key is stored encrypted in the MAXsystem vault, so an agent calls /ppt-from-text with a scoped token rather than handling the raw key.
Which MagicSlides endpoint should I use today?
Use POST /ppt-from-text — it accepts arbitrary input text and is the current supported generator. The /ppt_from_topic, /ppt_from_summery and /ppt_from_youtube endpoints are still in the spec but are marked deprecated, so new integrations should default to /ppt-from-text and only fall back to the others when their specific behaviours are required.
What are the rate limits for the MagicSlides API?
Per-endpoint rate limits are not declared in the OpenAPI spec; MagicSlides typically applies plan-tier ceilings tied to the API key. Check the plan attached to your key on magicslides.app for production volume planning.
How do I generate a presentation from text through Jentic?
Run pip install jentic, search for 'generate a presentation from text', and Jentic surfaces POST /ppt-from-text with its input schema. Execute with the source text, slide count and language fields, and persist the returned deck or download URL.
Can MagicSlides generate decks in languages other than English?
Yes. The API supports more than 100 languages — set the language field on the input payload to /ppt-from-text (or any of the legacy endpoints) and MagicSlides generates the deck in that language. The same generation endpoints are used regardless of language.