How codewithutkarsh12/peakcut-x works
PeakCut is a self-hosted service for turning uploaded or linked videos—and live streams—into captioned vertical clips, with optional social posting. The main workflow starts in the dashboard and server, then processes a video through transcription, clip selection, and rendering; live sessions add rolling recording and highlight detection. Separate branches cover creator workspaces, accounts, posting, and administration. Internal edges use the SOURCE INDEX references and excerpts; broader product-flow edges are supported by the README, so unsampled implementation details are not assumed.
Main entry points: jobs.py Video jobs · server.py HTTP server · live.py Live sessions · brain.py Clip selection
Components 24
Dashboard and access
- Web dashboard
web/index.html - HTTP server
studio/server.py - Sign-in and sessions
studio/auth.py - User accounts
studio/accounts.py
Video clipping
- Video jobs
studio/jobs.py - Video ingest
studio/media.py - Speech transcription
studio/transcribe.py - Clip selection
studio/brain.py - AI providers
studio/ai.py - Clip rendering
studio/render.py
Live clipping
- Live sessions
studio/live.py
Creators and posting
- Creator workspaces
studio/creators.py - Stream profiles
studio/profiles.py - Social posting
studio/posting.py
Platform services
- Data storage
studio/storage.py - Account and usage database
studio/db.py - User context
studio/ctx.py - Load management
studio/load.py - AI usage tracking
studio/usage.py - Update notices
studio/updates.py
Other components
- User
- Social platforms
- AI service
- Video or stream source
Connections 30
- User → Web dashboard: uses
README.md - Web dashboard → HTTP server: sends requests
studio/server.py - HTTP server → Sign-in and sessions: handles sign-in
studio/server.py - HTTP server → Video jobs: routes clip jobs
studio/server.py - HTTP server → Live sessions: routes live sessions
studio/server.py - HTTP server → Social posting: routes posting
studio/server.py - HTTP server → Creator workspaces: routes workspaces
studio/server.py - Video jobs → Video ingest: gets video
studio/jobs.py - Video jobs → Speech transcription: transcribes audio
studio/jobs.py - Video jobs → Clip selection: selects moments
studio/jobs.py - Clip selection → AI providers: requests analysis
studio/brain.py - AI providers → AI service: calls provider
README.md - Video jobs → Clip rendering: renders clips
studio/jobs.py - Video ingest → Video or stream source: downloads or probes
README.md - Live sessions → Video or stream source: records stream
README.md - Live sessions → Speech transcription: transcribes live audio
studio/live.py - Live sessions → Clip selection: selects highlights
README.md - Live sessions → Video jobs: creates clips
studio/live.py - Live sessions → Creator workspaces: uses workspace settings
studio/live.py - Creator workspaces → Stream profiles: uses profiles
studio/profiles.py - Social posting → Social platforms: publishes clips
README.md - Social posting → User accounts: uses connected accounts
studio/posting.py - Video jobs → Data storage: stores job data
README.md - Creator workspaces → Data storage: stores workspace data
studio/creators.py - Sign-in and sessions → Account and usage database: stores sessions
studio/auth.py - User accounts → User context: scopes user data
studio/accounts.py - Video jobs → Load management: checks capacity
studio/jobs.py - Clip selection → AI usage tracking: records AI usage
studio/brain.py - AI usage tracking → Account and usage database: writes usage records
studio/usage.py - HTTP server → Update notices: shows notices
studio/server.py