ShortGPT Video Automation
Automate scripts, voiceovers, asset sourcing, captions, and editing for short-form video production.
- Source repo
- RayVentura/ShortGPT
- Stars
- ★ 8.1k
- Last updated
- 1y ago
- License
- MIT
- Primary language
- Python
- FA score
- Insufficient evidence
At a glance
- How it runs
- Works with
- Platform-specificOpenAI API (Partial support)
- Cost
- Free software; you pay for model usage
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- A YouTube Shorts operator who wants to combine script generation, voiceover, asset preparation, and rendering in one automated workflow.
- Not a fit if
- Users unwilling to set up Docker, a .env file, and multiple third-party API keys
- Production teams needing mature docs and stable API examples
- Source review
- Insufficient evidence 8 safety controls not found
What does this agent do, and when should you use it?
ShortGPT is an MIT-licensed AI video-content automation framework for YouTube Shorts, TikTok, and related production workflows. It exposes ContentShortEngine, ContentVideoEngine, ContentTranslationEngine, and an EditingEngine based on Editing Markup Language and JSON. The project can run locally through Docker with a Gradio web interface at localhost:31415, and it also links to a Google Colab option. Its documented technology stack includes OpenAI, ElevenLabs, EdgeTTS, Pexels, Bing Image, and MoviePy, so adoption carries dependencies on those services and runtime components.
ContentShortEngine handles short-video creation from script generation through final rendering and includes YouTube metadata addition. ContentVideoEngine is intended for longer videos: it generates audio, automatically sources background footage, times captions, and prepares background assets. ContentTranslationEngine accepts a video file or YouTube link, transcribes audio, translates the content, voices it in a target language, adds captions, and returns a video in another language. EditingEngine uses Editing Markup Language and JSON to divide editing into customizable blocks; the framework also describes sourcing images or video through the web and Pexels API, and persisting automated-editing variables with TinyDB.
- A YouTube Shorts operator who wants to combine script generation, voiceover, asset preparation, and rendering in one automated workflow.
- A long-form video team that needs automated audio generation, background-footage sourcing, and caption timing.
- A localization producer who needs to dub and translate an existing video file or a YouTube link into a target language.
- A creator who wants multilingual voiceovers through EdgeTTS for languages such as English, Spanish, Arabic, Chinese, and Japanese.
- A developer who wants editable video-production steps represented as JSON and Editing Markup Language blocks.
How do you install or deploy this agent?
The README identifies Docker as the prerequisite for local use and directs readers to installation-notes.md, whose contents are not supplied. From the repository root, run:
docker build -t short_gpt_docker:latest .
docker run -p 31415:31415 --env-file .env short_gpt_docker:latestA Google Colab entry point is also linked. The README requires a .env file but does not document variable names, required credentials, Python version, or complete dependency setup, so credential configuration cannot be completed from the supplied material alone.
How do you use this agent?
After starting the container, open the Gradio interface at http://localhost:31415. For translation and dubbing, the README states that ContentTranslationEngine accepts a video file or YouTube link and returns a video with target-language voiceover and captions. The supplied material does not document web-form fields, Python API calls, or CLI arguments, so those exact interactions cannot be verified.
What are this agent's strengths and limitations?
- Covers three explicit production paths: short videos, longer videos, and video translation.
- ContentTranslationEngine documents an end-to-end path from a video file or YouTube link to transcription, translation, voiceover, captions, and a new video.
- EditingEngine represents editing in customizable Editing Markup Language and JSON blocks intended to be understandable to large language models.
- Offers both Docker-based local execution and a linked Google Colab option, with a Gradio web interface.
- The documented automation process uses OpenAI, and the README does not document a configuration path for replacing the model provider.
- Local deployment requires Docker and a .env file, but the supplied material does not list environment variables, credentials, or complete installation details.
- Asset sourcing depends on internet access, the Pexels API, and Bing Image, making operation dependent on external-service availability and access conditions.
- The README says more documentation is forthcoming and provides no verifiable web-interface walkthrough, API examples, or CLI parameters.
How does this agent compare with similar options?
The README lists both ElevenLabs and EdgeTTS for voice synthesis, and describes EdgeTTS as free with support for more languages than ElevenLabs.
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| ShortGPT Video Automation This agent | Insufficient evidence | Self-hosted serviceFree + model costs | ★ 8.1k | 1y ago | Python | — |
| Hypit Video Workflows | 71 · Some gaps | Agent plugin / skillFree + model costs | ★ 28k | 1d ago | TypeScript | Codex · Claude Code |
| Wonda CLI | 56 · Major gaps | CLIPaid | ★ 159 | 6d ago | TypeScript | Codex · Claude Code |
| Publora Social Publishing Skills | 67 · Some gaps | Agent plugin / skillFreemium | ★ 67 | 1d ago | Python | Codex · Claude Code · Claude.ai |
How does FollowAgents rate this agent?
Why each dimension lost points
Evidence only includes README-Docker.md, LICENSE, requirements.txt, and a GitHub Actions workflow. No permission model, user confirmation mechanism, data flow transparency, sensitive data handling policy, dependency security audit, external effects description, rollback mechanism, or source attribution. All criteria deducted for lack of evidence.
No code or tests to assess self-consistency, dependency availability, or failure messages. Dependency list exists but availability unverified. All criteria deducted for lack of evidence.
No documentation of target audience, use cases, capability boundaries, trigger precision, or environment fit. All criteria deducted for lack of evidence.
Information architecture unclear; install notes only for Docker and incomplete; naming stability not assessed; no examples or FAQ; no known limitations; license present but unverified; no versioning or changelog; maintenance responsibility unclear. All criteria deducted for lack or partial evidence.
No evidence of output usability, marginal value, or cost-benefit. All criteria deducted for lack of evidence.
No evidence of claim traceability, cross-source corroboration, or fact-inference separation. All criteria deducted for lack of evidence.
- Not found in source: least-privilege scopingGrant only what the task needs: a dedicated account or read-only token, scoped to specific directories and repos.
- Not found in source: confirmation before actingTurn on (or add) a confirmation step before it acts, and try it in a sandbox or test environment before real data.
- Not found in source: data-flow disclosureWatch which external services it contacts (proxy or firewall logs) and keep sensitive data out until you know where it goes.
- Not found in source: sensitive-data handlingUse dedicated, low-privilege, revocable API keys — never production credentials — and keep secrets out of logs.
- Not found in source: dependency securityPin versions and run a dependency audit (npm audit, pip-audit) before installing; prefer running it in a container.
- Not found in source: disclosed external effectsEstablish which external systems it writes to, sends to or changes, and verify with test accounts or repos before production.
- Not found in source: rollback or recovery pathBack up first, or work on a git branch or snapshot, so its changes can be undone.
- Not found in source: verifiable attributionInstall from the official repo or registry and check the publisher and URL to avoid look-alike packages.
- Repository contains only a few files, insufficient to support a comprehensive assessment of the Agent product.
- Dependency list includes pinned and unpinned versions, with no security audit provided.
- Docker run requires API keys, but no secure handling of keys is described.
Recent releases
Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.
-
v0.3.0 v0.3.0 - Gemini API Support
Release notes excerpt
ShortGPT Release v0.3.0 This release introduces support for the Gemini API, providing an alternative to OpenAI for content generation within ShortGPT. This allows users to leverage the capabilities of Gemini, especially in scenarios where it may offer advantages in cost, speed, or specific task performance. Major Changes: * Gemini API Integration: Added support for using the Gemini API as a primary or fallback option. Users can now configure their Gemini API key and utilize it for content generation tasks. …View on GitHub ↗ -
v0.2.0 v0.2.0 - Bug Fix
Release notes excerpt
ShortGPT Release v0.2.0 This release marks an overhaul and stabilization of ShortGPT after a period of inactivity. I've addressed core issues, upgraded key components, and made several quality-of-life improvements. Major Changes: * Stability Fixes: This release addresses numerous bugs that have accumulated, resulting in a much more stable and reliable experience. * GPT Model Upgrade: Migrated from OpenAI's gpt-3.5 to gpt4-o-mini. This brings improved performance, more accurate content generation, and better overall results. …View on GitHub ↗ -
Release notes excerpt
CHANGES Version 0.1.3 - Requiring a youtube url as the subscribe animation url in the EditingStep.ADD_SUBSCRIBE_ANIMATION step. - Adding a default subscribe animation youtube link by default shipped in the AssetDatabase - Making path imports relative for gpt prompts and editing blocks and flows. Version 0.1.2 - Improving logs in content engines Version 0.1.1 - Adding AssetType in AssetDatabase - Adding ApiProvider in api_db - Fixing pip libary missing editing_framework module, prompt_template module Version 0.1.0 - Fixing the AssetDatabase when it's empty Version 0.0.2 …View on GitHub ↗