Overview · Editor · AI and privacy
Install BridgeClip, create clips, follow jobs, and manage your Library and publishing queues. These instructions describe the current source; check the release notes for the features in your installed version.
Choose your platform on bridgeclip.ai or GitHub Releases. Packages include Python, FFmpeg and yt-dlp.
- macOS (Apple silicon or Intel): open the matching DMG and drag BridgeClip to Applications. Official builds are Developer ID signed and notarized by Apple.
- Windows x64: run the signed EXE installer.
- Linux x64: install the DEB with your package manager, or make the AppImage executable before opening it. An unlocked desktop secret service is required to save API keys.
See release verification for signatures and checksums.
BridgeClip keeps itself up to date. It checks Releases shortly after launch and every four hours, downloads a new version in the background, and installs it when you choose Restart to update (in the sidebar or Settings → About) or the next time you quit. macOS only installs an update signed by the same developer, and every download is checked against the SHA-512 published with the release.
Copies run from source, local package builds and apps opened straight from the disk image don't update themselves; Settings → About says why. To turn updates off, start BridgeClip with BRIDGECLIP_DISABLE_AUTO_UPDATE=1.
- Add an OpenRouter API key with available credit in the setup card. A saved key is never shown again; paste a new one to replace it, or choose Remove key in Settings.
- Open Create and use the file picker for a local video, or paste a public YouTube or completed Twitch VOD link. Dropping a local file opens the picker so you can grant access.
- Choose Automatic for finished exports, or Review & edit to adjust candidates in the editor.
- Choose format, framing, clip lengths, models and caption style, then generate.
- Follow progress in Jobs. Finished clips appear in Library and your output folder (by default,
~/BridgeClip).
Keys are encrypted with your operating system's secure storage. If secure storage is unavailable, BridgeClip asks you to configure or unlock it before saving keys.
Only download or clip material you have permission to use. Remote sites may limit downloads or change access rules.
In Create, choose a workflow before continuing: Automatic · Beginner friendly finds and exports clips; Review & edit · For advanced users lets you choose and adjust candidates before export. Each new video starts with neither workflow selected. YouTube links show a video card with its title, channel and thumbnail, plus duration, views and upload date when available. Metadata loads in the background and never blocks clip setup.
Review & edit always uses Jev candidate reviews. For Automatic, Jev review is off by default. Reviews use OpenRouter credit; see review behavior and costs.
In Create → Clips, choose Quality, Economy, or Advanced. Advanced offers searchable OpenRouter model pickers for transcription and clip planning, with model IDs, planning prices and compatibility notes. Both selections are required and appear in Review. Advanced retries the selected models without automatically switching models. Transcription must provide word timestamps; planning must support structured output. See model selection and transcription.
Smart framing follows faces and arranges screen shares with facecams shot by shot. Optional AI vision checks help with ambiguous layouts and can add OpenRouter cost. Use Review & edit to adjust crops and camera changes before exporting.
Choose from nine caption styles: Viral, Hormozi, Bold, Clean, Minimal, Fire, Glow, Neon and Karaoke. In Create → Captions, the selected style plays an animated sample showing how words appear and highlights advance. Pause or restart it to compare styles.
In Create → Format → Video speed, choose 1× (normal), 1.1×, 1.25×, 1.5×, 1.75×, or 2× for every clip in the job. Exports preserve voice pitch and keep captions synchronized. Speed works with Cut dead air and appears in Review and the saved results. Clip lengths and source trim times refer to the original footage: a 60-second clip at 1.5× exports in about 40 seconds, before any dead-air cuts. The choice stays selected when you clip another video in the same session.
Existing exports stay as they are; generate a new job to change their speed.
See video speed for timing and export details.
Paste a public, completed Twitch video link such as https://www.twitch.tv/videos/1234567890 into Create, then choose your clip settings and generate. BridgeClip downloads the saved video and uses the same transcription, AI moment selection and rendering flow as other sources. Links on twitch.tv, www.twitch.tv, m.twitch.tv and go.twitch.tv are accepted and normalized to the canonical video URL.
Live channels, Twitch clips, collections, subscriber-only videos and deleted or expired VODs are not supported. No Twitch login or cookies are used. The original source must be at most six hours and 20 GB. BridgeClip downloads the full source before applying the optional start and end times; a link's timestamp or tracking parameters are ignored. For a longer source, trim a downloaded file before adding it. Downloads also stop after four hours or when less than 1 GB of free space would remain.
Every run gets its own folder. The Library shows completed clips with virality scores, timecodes and tags. Jobs shows what is running or queued right now (up to two clipping runs go at once; more wait in a queue) and every earlier run, including completed, failed, cancelled and interrupted jobs; completed runs open directly in Library, and failed runs from this session can run again.
Each previous job has a three-dot menu for Open in Library (or Open job for session jobs), Open folder and Details. Details opens the saved transcript and edit trace, including for failed runs. Older runs without a saved status appear as unfinished. You can optionally connect social accounts through Zernio to publish or schedule a selected clip.
During creation, each stage shows its own elapsed time and progress: downloaded bytes, transcription chunks, prepared candidates, rendered clips, saved files and editor-preview encoding where measurable. AI requests without measurable progress show an indeterminate bar. The overall percentage is an estimate; completed runs retain a Processing time by stage breakdown in the Library and editor. Each stage keeps its own color in the time breakdown. The progress page reuses the source preview from Create for YouTube and local files.
New runs show live model requests, input/output tokens and provider-reported cost; missing usage stays unknown and partial totals exclude unreported charges. Inside frame & review separates face sampling, camera scans, detailed face tracking, shot-layout checks and Jev review, with the current candidate and accumulated timings. These diagnostics are saved with the run and remain available in its processing-time details. See the framing performance investigation for the long-video decoding fix.
Inside a Library run, Search clips filters by title, tag or clip number across both posted and unposted clips. Selection and bulk actions apply only to visible results. The YouTube icon beside Inspect transcript & edits opens the original video in your browser when a YouTube source link is saved. Not Posted and Posted are independently collapsible sections with clip counts. Not Posted opens by default and appears first; Posted starts collapsed below it. Selection applies only to expanded sections.
Badges distinguish scheduled, publishing, partially posted and failed posts using this workspace’s saved posting history.
Each clip’s … menu contains posting, adding to an automation, opening its folder, and Delete clip. Delete a single clip directly from this menu without selecting it first; the confirmation names the clip before permanently removing its local export. Choose Mark as posted for clips you published yourself; the mark persists without a connected account and can be undone from the same menu. It changes local Library status only.
Select clips to reveal the red trash button for deleting just those exports, with confirmation; source footage, editor edits and other clips are preserved. The post dialog can enhance platform-specific titles, captions and tags using the same transcript, source context and optional research as Automations; review and apply the draft before posting.
Library overview cards show the total clip count at the top left and separate Posted / Not Posted counts below the title. Bookmark a run to keep it in the Bookmarked section. Cards move smoothly between Bookmarked and Recent runs without reloading their thumbnails; removing the last bookmark returns to one All runs grid. Bookmarks persist across restarts, and reduced-motion preferences turn off movement.
The trash action asks for confirmation before permanently deleting the run folder, all clips and other files inside it, and cached clip previews. It does not remove published social posts or copies stored elsewhere. Active runs cannot be deleted.
Settings → About → Content storage shows the total file size and file count in your output folder. Refresh to recalculate; inaccessible files are reported as a partial total.
Connect social accounts with your own Zernio account and API key to post or schedule clips. Metadata enhancement uses your OpenRouter key. For platform-specific drafts, scheduling consent and required TikTok reviews, see publishing and metadata. BridgeClip must be open for daily automations to run.
In Automations, drag a queued clip’s handle to change its order with animated movement and automatic scrolling. For keyboard sorting, focus the handle, press Space to pick up the clip, use the Up/Down arrow keys, then press Space to drop or Escape to cancel. Submitted or uncertain items stay in place. The bank separates Queued, Needs attention (when needed), and Submitted clips. The Queued heading shows the next posting slot and its timezone, or indicates that scheduling is paused or unconfigured. Hover or focus each row’s info icon for its added/submitted timestamps, source, caption and metadata status.
Use the actions menu to edit a bank item, Show in Finder (Show in folder on other systems) to reveal its saved video, or Remove from queue, and View in Library to open its source run filtered to that exact clip, including submitted clips and items with enhanced titles. Choose Show all clips to return to the full run. Submitted clips have no removal option. Removing a queued clip leaves the original file intact.
Reviewed metadata is marked Enhanced in the info tooltip and hides the Enhance button; selecting a platform that still needs metadata makes enhancement available again.
Enhance by source video works with both linked videos and attached files. Select a source and optionally describe the video or suggest an emphasis for its titles, captions and tags. Review the generated draft before applying it.
In Automations, the header shows scheduling status once. More settings holds caption writing, YouTube options and the automation name; previous failures are expandable historical messages. Use Acknowledge warnings for one automation or Acknowledge all warnings for every automation to clear existing failure indicators without a successful post. Acknowledgements survive restarts; new failures warn again. Messages remain in the run details or clip info, and acknowledged clips awaiting review appear under Held clips without being retried or returned to the queue.
Dismiss a clip’s upload or metadata error with its × button; dismissal survives restarts, preserves the message in clip info, and a new failure shows the warning again. Run now prioritizes ready clips without previous upload errors. Use Retry clip to retry a specific failed upload without changing queue order, even after dismissing its error. Retry is available only for queued clips that have not been submitted; metadata drafts and required TikTok review must be resolved first.
Held clips have a visible Return to queue action and, when linked to a Zernio post, Refresh post status. A fresh Zernio check moves already published or active posts to Submitted and permits requeuing only fully failed posts; partial or uncertain results stay held for review in Posts. Returning a failed linked clip disables Retry on its old Posts entry so only the new queue attempt can post.
Recovery also works when local post history was removed or aged out, provided Zernio returns complete post details. Returning an unlinked clip requires confirming it was not published. Required setup and TikTok approvals still apply.
Check Zernio status makes a fresh, read-only account health request and shows connection/permission issues without uploading or posting. Zernio’s documented health API does not expose its protective upload cooldown, so the result explicitly leaves that hold unconfirmed even when the connection is healthy. Failed checks never fall back to cached success.
- Run Settings → System check and use the in-app error to identify missing dependencies or provider issues. In development, set Python path if the virtual environment is not detected.
- If a link fails, check it in a signed-out browser or download the video yourself and select the local file.
- Open a job’s Details to inspect its saved transcript and edit trace, including failed runs. Inspecting saved results makes no provider calls. See saved reviews and diagnostics.
- Logs omit raw provider responses and private source details, but review logs and run files before sharing them. See local storage and cleanup.
For help, use GitHub Issues or Discord. Report security issues using SECURITY.md.