AI tools, scored for your job
Get paid to train AI
🤝 You're in the Free & Open Source library — community-built tools you can use without paying. A different concept from our scored AI directory.
Home › Free & Open Source › HR › subtitle-scout
✦ Free & open sourceHR

subtitle-scout

An LLM agent-driven subtitle finder, evaluator, and translator for local media.

AI-Powered Subtitle Automation Agent

Autonomous LLM agent that finds, validates, installs, and translates media subtitles.

Free forever · open source · AGPL-3.0
✓ Needs IT✓ Browser-based✓ Real-time collaboration
6.9
ZEKAI
Precise media subtitle automation
Free & open source · HR
Ease of setup5
Usefulness8
Maintenance9.5
Community5
★ 125
GITHUB
Sep 2026
UPDATED
AGPL-3.0
LICENSE
🛠 Don't want to self-host? We can set it up →
⚡ Quick answer

Subtitle Scout is a self-hosted subtitle management system driven by an LLM agent. It targets media collectors and engineers managing local video libraries who want automated, context-aware subtitle matching rather than rigid pipeline searches.

Runs on
Web
Hosting
Self-host
Languages
TypeScript
License
AGPL-3.0

What it is

Instead of relying on top search hits, the tool's core agent downloads subtitle candidates into a sandbox, checks timeline spans and cue counts, and evaluates structural fit for exact episodes. When no source exists, a secondary agent handles sentence-by-sentence translation using custom glossaries.

Who it's for

It suits power users and technical media managers tired of mismatched release groups or missing translations. It does not suit users looking for a simple point-and-click desktop app without Docker or API keys.

Key features

LLM Episode Verification

Downloads candidates into a sandbox to judge exact episode fit based on timeline span and cue structure.

Multi-Source Aggregation

Queries OpenSubtitles, SubDL, ASSRT, r3sub, Jimaku, SubHD, and Zimuku depending on your target language.

Agentic Translation Loop

Translates subtitles sentence by sentence using a glossary pass when no source tracks are available.

Direct Folder Scanning

Scans your media directories directly without requiring Plex, Jellyfin, Emby, Sonarr, or Radarr.

How to start

Up and running in 3 steps

1
Clone the repository and create your local cache and environment file.
2
Start the application stack using Docker Compose.
3
Open the web wizard on port 8099 to configure your TMDB, subtitle source, and LLM API keys.
Real talk

Our honest verdict

What's great

  • Evaluates subtitles based on actual content rather than blindly trusting release group metadata
  • Translates missing languages using a dedicated agent loop with glossary support
  • Maintains an independent SQLite database with local credentials instead of third-party sync
  • Includes an Agent Skill file to let coding agents handle deployment setup

Where it falls short

  • Requires active API keys for TMDB, translation, and multiple subtitle providers to function fully
  • Requires manual Docker and environment configuration on a self-hosted server
  • Weaker LLM tiers will hallucinate match results instead of failing cleanly
Cost

Pricing & license

Free under its open source licence. The only cost is running the hardware and paying for the LLM and API tokens you consume.

Paid work

Open remote AI-training roles

880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.

See all open roles →
Compare

Alternatives

🎁 Free & open-source alternatives
Get more

Use it, or have us set it up

📄 Free PDF: subtitle-scout for HR
A plain-language guide to getting your first result. Sent to your inbox.
🔔 64 people want a full video tutorial — add your vote
🛠 Don't want to set it up yourself?
We'll install, host & configure subtitle-scout for you — and train your team.
Get it done for you ↗
Questions

FAQ

Does this require Plex or Sonarr?

No, Subtitle Scout scans your media folders directly and operates entirely independently.

Which languages are supported?

Chinese has the deepest coverage, with full support for English, Japanese, and Korean, plus basic support for several European languages.

What happens if a model tier is too low?

Models below frontier mini or flash tiers tend to fabricate match results, leading to misidentified media.

Radar

Latest AI news

✉️ The weekly free toolkit
The best new free & open-source tools for HR — one short email a week, in your language.
Free · 1 email/week · unsubscribe anytime
Get paid to train AI →

See Zekai first in Google