Changelog

Every Podcli release and what changed. Notes are pulled straight from the GitHub releases.

Releases on GitHub
  1. v2.3.4Latest
    Highlights

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.3.0...v2.3.4

  2. v2.3.2
    Highlights

    Patch fixes for the clip suggestion and transcript flows.

    Fixed

    • Suggestions tell you what actually failed. When clip suggestion can't produce results, it now surfaces the real reason (not logged in, usage or rate limit reached, timeout, or an unexpected crash) instead of the generic "check claude/codex login" message.
    • Suggestions stop repeating moments. Suggestion now knows about clips already rendered for the current video plus the moments already on screen, and skips those ranges.
    • Long clip titles are fully visible. The clip title editor wraps instead of scrolling a single-line box.

    Changed

    • The Whisper / AssemblyAI engine choice now persists with your saved preset, so a show keeps its preferred transcription engine.

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.3.1...v2.3.2

  3. v2.3.1
    Highlights

    Patch fixes for the studio.

    Fixed

    • Suggestions no longer repeat moments. Clip suggestion now knows about clips you have already rendered for the current video, plus the moments already on screen, and skips those ranges instead of proposing them again.
    • Long clip titles are fully visible. The clip title editor wraps to multiple lines instead of scrolling a single-line box.

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.3.0...v2.3.1

  4. v2.3.0
    Highlights

    Highlights

    A new Highlights studio turns any long video into short clips without scrubbing. Point it at a file and it finds the standout moments on its own, then lets you fine-tune and export.

    Highlights studio (new)

    • Auto-detects the best moments from laughter, cheering, energy and motion. No need to pick a content type.
    • Trim each moment on a video timeline with draggable in and out handles. Edits re-cut in seconds.
    • Render vertical (9:16), horizontal (16:9) or square (1:1).
    • Download the whole reel or individual clips.
    • Save, reopen and delete sessions. Drop or browse a video to start.

    Moment detection engine

    • Laughter and reaction detection via YAMNet (ONNX), on the no-torch path so it runs in the native runtimes.
    • Saliency-based detection for party and action footage, plus a generalized auto profile.
    • Pool detection across a folder of clips.
    • Windowed inference so long videos never block the worker.

    Transcription

    • AssemblyAI added as a selectable transcription engine.

    Windows

    • Uninstall cleans up more completely.

    CLI and MCP

    • podcli reel gains --format, --min-dur, --max-dur, plus list and delete.
    • manage_reel and detect_highlights exposed over MCP.

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.2.1...v2.3.0

  5. v2.2.1
    Formats & captionsContent studio

    ⚠️ Behavior change: podcli uninstall now removes everything by default

    podcli uninstall previously kept your user data (config, knowledge, presets, assets, history, cache) unless you passed --purge. It now removes the entire managed podcli folder, including that user data, by default.

    • If you want to keep your data, back up your podcli folder before uninstalling, or run podcli uninstall --dry-run first to see exactly what will be removed.
    • --purge is now a no-op alias, kept only for compatibility (uninstall already removes everything).
    • Uninstall still prompts for confirmation unless you pass --yes.

    Changes

    • Native podcli uninstall removes the full managed podcli folder by default (runtime files, models, cached data, and the managed user-data folder). (#41)
    • Windows: uninstall now removes %LOCALAPPDATA%\podcli\bin from the user PATH, and --dry-run reports the PATH cleanup it would perform.
    • install.ps1 gains -Uninstall -Purge support with the same PATH cleanup.
    • Uninstall tests updated for the new default behavior.

    Full changelog: https://github.com/nmbrthirteen/podcli/compare/v2.2.0...v2.2.1

  6. v2.2.0
    Formats & captionsContent studio

    Highlights

    • Horizontal (16:9) and square (1:1) clips alongside vertical 9:16, with per-format captions that scale to the canvas instead of floating mid-frame. Set the format in the studio (New episode → Settings → Format), via the create_clip / batch_create_clips MCP tools, or with podcli process --format. Defaults to vertical, so existing behavior is unchanged. (#37)
    • Content studio — free-form "ask for anything" generation (LinkedIn post, thread, extra hooks, …) plus per-section Regenerate for titles, description, tags, and hashtags with optional guidance, and a live thumbnail preview. (#37)
    • YouTube / direct-URL import — paste a link into the studio's video field to download it via yt-dlp and continue straight into the preview/transcription workflow. (#40)
    • Thumbnail CLI flags — --thumbnails / --no-thumbnails on podcli process to force thumbnail generation on or off. (#38)

    Studio UI

    • Refreshed palette and a consistency pass: one low-opacity hairline border system, darker surfaces, depth from shadow, all-SVG icons, portaled modals, a sticky preview pane, tabular figures, and loading buttons that stay legible.
    • Episode work now survives navigation — export results, energy analysis, suggestions, and selections persist when you leave and return to the page.

    Under the hood

    • New FormatSpec single source of truth (backend/services/formats.py) replaces scattered 1080×1920 hardcodes; format is threaded end to end with format-aware duration caps, and the render path forks on the format's reframe strategy (face-tracked crop for vertical/square, scale-and-pad letterbox for horizontal).

    Full changelog: https://github.com/nmbrthirteen/podcli/compare/v2.1.2...v2.2.0

  7. v2.1.2
    Analytics

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.1.1...v2.1.2

  8. v2.1.1
    Analytics

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.1.0...v2.1.1

  9. v2.1.0
    Analytics

    What's Changed

    New Contributors

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.13...v2.1.0

  10. v2.0.13

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.12...v2.0.13

  11. v2.0.12

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.11...v2.0.12

  12. v2.0.11

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.10...v2.0.11

  13. v2.0.10

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.7...v2.0.10

  14. v2.0.9

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.8...v2.0.9

  15. v2.0.8

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.7...v2.0.8

  16. v2.0.7

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.6...v2.0.7

  17. v2.0.6

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.5...v2.0.6

  18. v2.0.5

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.4...v2.0.5

  19. v2.0.4

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.3...v2.0.4

  20. v2.0.3

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.2...v2.0.3

  21. v2.0.2

    What's Changed

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v2.0.1...v2.0.2

  22. v2.0.1

    What's Changed

    New Contributors

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v1.1.0...v2.0.1

  23. v2.0.0

    What's Changed

    New Contributors

    Full Changelog: https://github.com/nmbrthirteen/podcli/compare/v1.1.0...v2.0.0

  24. v1.1.0

    v1.1.0 — Studio Dashboard

    podcli v1.1.0 — Studio Dashboard

    A full content studio on top of the podcli engine: browse and re-iterate on rendered clips, fix framing visually, and learn from YouTube performance. Backward compatible with v1.0.0 (additive data fields, new tools/routes).

    Studio dashboard (web UI)

    • Vite + React Router app with a sidebar shell; clip library with a card grid and a custom clip player
    • Redesigned clip detail page (player hero + grouped editor)
    • Per-clip thumbnail studio: text/line control, keyframe frame-pick, image upload, variation picker, full brand-template editor
    • Delete clips end-to-end; live export indicator as clips stream in while rendering

    Reframe / crop pipeline

    • Manual reframe editor to fix camera framing on rendered clips — visual timeline, play, jump-to-keyframe, drag-scrub, visual trim
    • Manual keyframe crop mode and camera-switch detection in the render pipeline
    • Keyframes + trim persist across reopen; full render recipe replayed so brand survives re-render

    YouTube analytics learning loop

    • Phase 2: performance loop that feeds back into clip-picking; Phase 3: semantic "what separates top performers" analysis
    • Web clip-linking for the live OAuth path (no longer CLI-only); CSV import path needs no auth
    • Hardened credentials (integrations.json 0600, refresh-token validation); transparent CSV attribution with match scores

    Engine / quality

    • New CLI subcommands: clips list/edit/reopen/delete, youtube auth/status/link/sync/learn
    • Atomic history writes + serialized mutations; structured crop-strategy logging; shared API client, error utilities, and asset resolver
    • Deduplication and hardening pass across the dual-server + CLI codebase

    Full changelog: https://github.com/nmbrthirteen/podcli/compare/v1.0.0...v1.1.0

  25. v1.0.0

    First stable release of podcli — transcribe, clip, and publish short-form video from one place.

    License: podcli is now AGPL-3.0 (changed from MIT in this cycle). If you distribute or run a modified version as a network service, the AGPL terms apply.

    Highlights

    Config profiles & portable setup (#1)

    • Split config home (.podcli/ — knowledge, presets, assets, corrections, integrations) from the data dir (data/ — cache, transcripts, renders), overridable via PODCLI_HOME / PODCLI_DATA.
    • Export / import / activate config bundles to move your whole show setup between machines. Imported bundles no longer carry the exporter's absolute paths.
    • Auto-migration of legacy .podcli/cache and presets/ into the new layout, idempotent and non-blocking.

    Integrations plugin system (#1)

    • Pluggable, category-based integration registry exposed over MCP and the web UI, with enable/disable state.

    DaVinci Resolve export — beta (#1)

    • Export shorts as a Resolve-importable FCPXML timeline; each short becomes an editable compound clip with source on V1 and a ProRes 4444 alpha caption overlay on V2.
    • Emits canonical NTSC frame durations and DF/NDF timecode.
    • ⚠️ Beta: the FCPXML math is unit-tested but end-to-end import into Resolve 20.x hasn't been broadly verified yet. Please report issues with your Resolve version and source format.

    Reliability

    • Non-interactive / CI safe (#3): podcli process no longer crashes with OSError [Errno 22] when run with piped stdin or no TTY — it renders all suggested clips headlessly instead.
    • Resume caching: suggestion sessions are reused across runs, keyed on the inputs that affect selection (and skipped when a custom transcript is supplied).
    • Repo hygiene: CODEOWNERS for review routing (#2), frame-seam fix for split-screen thumbnails/crops.

    Notes

    • Requires ffmpeg (with libass for caption burn-in).
    • 299 Python + 41 TS tests passing; TypeScript typecheck clean.

    Full changelog: https://github.com/nmbrthirteen/podcli/compare/v0.1.0...v1.0.0

  26. v0.1.0

    v0.1.0 - Initial release

    Podcli is an open-source AI podcast clipper. Runs on your laptop. One command turns a long-form episode into vertical clips with face tracking and burned-in captions.

    What's in it

    • One-command pipeline - ./podcli process episode.mp4 does the whole thing
    • AI clip suggestion - Claude or Codex picks moments scored on 4 dimensions
    • Face tracking - YuNet detection with smoothed camera, handles split-screen layouts
    • 4 caption styles - branded, hormozi, karaoke, subtle (burned in)
    • Hardware-accelerated export - VideoToolbox on Mac, NVENC on NVIDIA, VAAPI on Linux
    • 3 interfaces - CLI, MCP server for Claude Code, drag-and-drop web app
    • Knowledge base - 13 markdown files teach the AI your show's voice
    • PodStack workflow - slash commands for titles, descriptions, thumbnails, publish checklists

    Install