Open-source async video messaging. Self-host it on your own infrastructure.
SendRec is an open-source alternative to Loom that you self-host on your own infrastructure. Record your screen, share videos with your team, and keep full control of your data. GDPR native, runs anywhere Docker runs.
| SendRec | Loom | Cap | Zight | |
|---|---|---|---|---|
| Open source | Yes (AGPL) | No | Yes (AGPL) | No |
| Self-hostable | Yes | No | Yes | No |
| Browser recording | Yes | Yes | No | Yes |
| EU data residency | Yes | No | No | No |
| Team workspaces | Yes | Yes | No | Limited |
| SSO / SAML | Yes | Enterprise | No | Enterprise |
| AI transcription | Yes (local or cloud) | $20/user | Paid | No |
| Per-viewer analytics | Yes | Yes | No | No |
| Free commercial use | Yes | N/A | No | N/A |
| Pricing | €0–12/month | $15–20/user | $12/month | $9–11/user |
Record your screen, camera, or both — directly in the browser. Works on Chrome, Safari, and Firefox. Records MP4 natively when supported, with automatic WebM fallback. A 3-2-1 countdown gives you time to prepare. Pause and resume, draw annotations live, flip between front and back cameras on mobile.
Drag and drop up to 10 videos at once — MP4, WebM, and MOV. Files upload directly to S3 via presigned URLs, and each gets a shareable link instantly.
Every video gets a shareable link with password protection, expiry dates, and per-video download controls. Trim videos, upload custom thumbnails, and add call-to-action buttons that appear when the video ends.
Server-rendered watch pages with a custom video player (seek bar, speed control, keyboard shortcuts, picture-in-picture), AI-generated summaries and chapter markers, a clickable transcript panel, closed captions, timestamped comments, and emoji reactions. iOS Safari gets native controls for reliable playback. SEO-optimized with OpenGraph tags, Twitter Cards, and VideoObject JSON-LD. Embed videos and playlists in docs and wikis with a lightweight iframe player.
Per-video analytics with view counts, completion funnel (25/50/75/100%), and CTA click-through rates. Get notified on views, comments, or as a daily digest.
- Screen & camera recording — works on Chrome, Safari, and Firefox; prefers MP4 with WebM fallback; 3-2-1 countdown, pause/resume, webcam overlay, drawing annotations, mobile front/back camera
- Video upload — drag-and-drop up to 10 files at once, MP4/WebM/MOV, per-file progress
- Automatic transcription — whisper.cpp (local) or OpenAI-compatible / Deepgram cloud providers, closed captions on watch and embed pages, full-text search
- Transcript editing — trim by clicking transcript segments, filler word removal with preview, AI-generated title suggestions
- Sharing — expiring or permanent links, password protection, per-video download toggle, custom thumbnails
- Comments & reactions — timestamped comments, emoji reactions, configurable modes
- CTA buttons — call-to-action overlay on video end with click tracking
- AI summaries — AI-generated summaries and chapter markers in the seek bar via any OpenAI-compatible API
- Viewer analytics — daily view charts, completion funnel, CTA click-through rates
- Generic webhooks — POST events (video created/ready/deleted, viewed, comment, milestone, CTA click) to any URL with HMAC-SHA256 signing, retries, and delivery log
- Slack notifications — per-user Slack incoming webhook for view and comment alerts
- View notifications — off, views only, comments only, both, or daily digest
- Embeddable player — lightweight iframe for videos and playlists, with captions, CTA, and milestone tracking
- Custom branding — logo, colors, footer text, custom CSS injection, per-user defaults with per-video overrides
- Library — folders, tags, and playlists; search by title and transcript; batch delete/move/tag; inline title editing
- Playlists — curated video collections with custom ordering, shared watch pages with auto-advance, watched badges, password protection, and email gating
- Dark/light mode — system preference detection, manual toggle, theme-aware charts
- Email gate — require viewer email before watching, per-viewer analytics with email, completion tracking
- SEO — OpenGraph tags, Twitter Cards, VideoObject JSON-LD, canonical URLs, robots.txt
- Subscription billing — optional Creem integration for free/Pro tiers, webhook-based plan activation, customer portal
- Integrations — Jira and GitHub issue creation, Nextcloud (oEmbed + API keys), per-user API keys, OpenAPI docs
- Team workspaces — shared video libraries with role-based access (owner/admin/member/viewer), email invites, workspace-level branding and billing, video transfer between personal and workspace scopes
- SSO — Google, Microsoft, and GitHub social login; workspace OIDC and SAML 2.0; SCIM 2.0 provisioning for automated user lifecycle management
- Self-hostable — single Go binary, Docker Compose, PostgreSQL, S3-compatible storage
Create a docker-compose.yml and run docker compose up -d:
services:
sendrec:
image: ghcr.io/sendrec/sendrec:latest
ports:
- "8080:8080"
environment:
- DATABASE_URL=postgres://sendrec:secret@postgres:5432/sendrec?sslmode=disable
- JWT_SECRET=change-me-to-a-long-random-string
- BASE_URL=http://localhost:8080
depends_on:
postgres:
condition: service_healthy
postgres:
image: postgres:18-alpine
environment:
POSTGRES_USER: sendrec
POSTGRES_PASSWORD: secret
POSTGRES_DB: sendrec
volumes:
- db-data:/var/lib/postgresql/data
healthcheck:
test: ["CMD-SHELL", "pg_isready -U sendrec"]
interval: 5s
timeout: 5s
retries: 5
volumes:
db-data:Open http://localhost:8080, register an account, and start recording.
For S3 storage, transcription, and production setup, see the Self-Hosting Guide.
SendRec also ships with a Helm chart in helm/sendrec for Kubernetes deployments.
Create your own values file such as values-prod.yaml and keep environment-specific settings there instead of editing the chart defaults in-place:
sendrec:
env:
baseUrl: "https://sendrec.yourdomain.com"
s3Endpoint: "https://s3.amazonaws.com"
s3PublicEndpoint: "https://s3.amazonaws.com"
s3Bucket: "your-bucket-name"
s3Region: "eu-central-1"
transcriptionEnabled: "false"
googleAuthAllowedDomains: "example.com"
secrets:
databaseUrl: "postgres://sendrec:secret@postgres:5432/sendrec"
jwtSecret: "change-me-to-a-long-random-string"
s3AccessKey: "your-access-key"
s3SecretKey: "your-secret-key"Install the chart:
helm install sendrec ./helm/sendrec \
--namespace sendrec \
--create-namespace \
-f values-prod.yamlPreview the rendered manifests before applying changes:
helm template sendrec ./helm/sendrec -f values-prod.yamlUpgrade an existing release after changing your values file:
helm upgrade sendrec ./helm/sendrec \
--namespace sendrec \
-f values-prod.yamlThe chart configures:
- A Deployment for the SendRec app
- A ConfigMap for non-secret environment variables
- A Secret for credentials and API keys
- A Service and optional Ingress
- An optional NetworkPolicy
- Optional whisper model storage and init container when
sendrec.env.transcriptionEnabled="true"
For production setup details such as reverse proxying, object storage, and operational guidance, see the Self-Hosting Guide.
git clone https://github.com/sendrec/sendrec.git
cd sendrec
cp .env.example .env
docker compose -f docker-compose.dev.yml up --buildPrerequisites: Go 1.25+, Node 24+, pnpm, Docker
# Run the full stack with Docker
make docker-up
# Or run services separately:
make dev-web # Frontend dev server (port 5173, proxies API to 8080)
make run # Go server (requires DATABASE_URL, S3 env vars)
# Build everything
make build
# Run tests
make test- Frontend: React 19, TypeScript 5.9, Vite 7
- Backend: Go (single binary, chi router)
- Database: PostgreSQL 18
- Storage: S3-compatible object storage (Garage for dev and self-hosting)
- Transcription: whisper.cpp (local default), or any OpenAI-compatible Whisper API (OpenAI, Groq, Scaleway, etc.), or Deepgram
- Deployment: Docker Compose
Single Go binary that:
- Serves the React SPA (embedded at build time)
- Handles REST API requests (
/api/*) - Serves interactive API documentation (
/api/docs) - Renders server-side watch pages with OpenGraph tags, Twitter Cards, and JSON-LD (
/watch/:token) - Runs database migrations on startup
Video recordings happen entirely in the browser using getDisplayMedia + MediaRecorder. The recorder prefers MP4 (native in Chrome 130+ and Safari) with automatic fallback to WebM for Firefox and older browsers. Users can also upload existing video files (MP4, WebM, MOV) via drag-and-drop. Files upload directly to S3 via presigned URLs — the server never touches video bytes.
After upload, the server generates a thumbnail with ffmpeg and enqueues the video for transcription. The transcription provider is selectable via TRANSCRIPTION_PROVIDER: local whisper.cpp (default), any OpenAI-compatible Whisper API, or Deepgram. WebM recordings are automatically transcoded to MP4 for universal playback. Uploaded MP4s are normalized with iOS-safe encoding flags when needed. Transcripts are stored as VTT subtitles and a clickable segment panel on the watch page. Transcription is optional — if TRANSCRIPTION_ENABLED=false, it is silently skipped.
SendRec can run either on a single server with Docker Compose or on Kubernetes with the included Helm chart. See the Self-Hosting Guide for full setup instructions, including:
- Production Docker Compose configuration
- Helm and Kubernetes configuration
- Environment variables reference
- Reverse proxy setup (Caddy example)
- Enabling transcription (whisper.cpp local, OpenAI-compatible, or Deepgram)
- Removing usage limits
- Updating and backups
Interactive API reference is available at /api/docs when running your instance (powered by Scalar). The raw OpenAPI 3.0 spec is at /api/docs/openapi.yaml.
SendRec is licensed under the GNU Affero General Public License v3.0.
- Self-Hosting Guide: SELF-HOSTING.md
- Changelog: GitHub Releases
- Website: sendrec.eu
- Email: [email protected]