Your podcast,
clipped for you
Podcast to
content studio

Podcli finds your best moments, crops them to vertical, and adds captions. Clips ready for TikTok, Reels, and Shorts. No editing skills needed.A local-first podcast CLI. Hand it an episode, get back vertical clips with face-tracked crop and captions burned in. Drive it from Claude Code, Codex, Cursor, or any MCP client. Just say “clip this episode.”

Watch the demo
Tech Daily
the real reason
The Founder
wait for this
podcli

Clips made to stop the scrollVertical clips, straight out of the pipeline

Crop, captions, brand. Ready to post.9:16, 16:9, or 1:1. Cropped and captioned locally.

Tech Daily
the one thing
The Founder
how we scaled
Deep Dive
that one call
Long Story
it gets wild

Mock previews. Your episodes, your speakers, your brand.

Everything you need to post more

Podcli does the editing busywork so you can stay on camera, not in the timeline.

One episode. A week of posts.

Upload once.

Best moments, found.

Skip the scrub.

Speaker stays framed.

Every format.

Captions, timed by word.

Four styles.

Your brand, baked in.

Logo included.

See what landed.

Make more of it.

Drop in. Clips out.

Browser studio.

Free and local.

No watermark.

The podcast CLI that runs from one binary

Everything is local. The only network call is the optional AI scoring step, and only if you bring a key.

Whisper transcription

Words and speakers.

Face-tracking crop

Every aspect ratio.

Burned-in captions

Word-level timing.

Studio dashboard

Preview, then export.

AI clip suggestion

Your rules included.

Performance loop

Learn from views.

MCP server, 26 tools

Agent-ready.

Hardware encoding

Fast where available.

From episode to posted, in three steps

No timeline. No editing software. No watermark.

01

Add your episode

Drop in a video file or point it at a recording

02

AI picks the moments

It finds the lines worth clipping and crops to vertical

03

Download & post

Captioned, branded clips ready for every platform

One command, six stages

CLI, Web UI, or AI agent. Same pipeline.

01

Transcribe

Whisper + pyannote for word timestamps and speaker labels

02

Score

Claude or Codex reads the transcript plus your knowledge base and picks clips

03

Export

Batch render 1080×1920 MP4s with captions and −14 LUFS audio

Show the content workflow: stages 04 to 06+
04

Titles & Descriptions

8 options per clip, descriptions with hooks and SEO keywords

05

Thumbnails

Two-line text briefs for podcast (16:9) and shorts (9:16)

06

Publish

Pre-upload checklist, first-24-hours ops, day 3 to 4 optimization

01-03: video engine. 04-06: PodStack slash commands.

Captions that keep people watchingFour styles, one flag

Pick a look. It stays consistent.Word-level timing. Four styles.

Your Show
the key to success is showing up every single day

branded--caption-style branded

The safe default.

Your Show
THE KEY TO SUCCESS IS SHOWING UP EVERY DAY

hormozi--caption-style hormozi

Bold and high contrast.

Your Show
the key to success is showing up every single day

karaoke--caption-style karaoke

Words light up as they land.

Your Show
the key to success is showing up every single day

subtle--caption-style subtle

Clean and quiet.

Plan before recording. Publish after. PodStack handles both.

Plan, make, publish. Your voice stays in the loop.Nine slash commands. One content pipeline.

01

Plan

Brief, arc, hooks

  • Episode Architect/plan-episode
  • Content Analyst/process-transcript
  • Brand Guardian/review-content
02

Make

Clips, titles, thumbnails

  • Producer/produce-shorts
  • Title Writer/generate-titles
  • Art Director/plan-thumbnails
03

Publish

Copy, launch, learn

  • Copywriter/generate-descriptions
  • Launch Manager/publish-checklist
  • Analyst/retro-episode
BR
Brand memory
Your show & hosts
Identity, positioning, who's on the mic
Voice & banned words
How you sound, what to never say
Title formulas
The title shapes that work for you
Thumbnail style
Layouts, colors, and type
.podcli/knowledge/
  01-brand-identity.md · show, positioning, hosts
  02-voice-and-tone.md · voice, banned words
  05-title-formulas.md · title shapes, rules
  07-thumbnail-guide.md · layouts, colors, type
  ... 13 files, all customizable

It's an MCP server

Give your agent an episode. It returns clips.

See all 26 MCP tools
01

Understand

transcribe_podcastimport_transcriptanalyze_energy
02

Choose

suggest_clipsmodify_clipknowledge_base
03

Render

create_clipbatch_create_clipsmanage_assets
Plus 17 support tools for jobs, settings, history, and export.
Show local MCP configuration+
// claude_desktop_config.json or codex setup
{
  "mcpServers": {
    "podcli": {
      "command": "podcli",
      "args": ["mcp"]
    }
  }
}

Questions

How does Podcli compare to hosted clippers?+

Podcli is a free, open-source, local-first podcast clip generator: Whisper transcription, AI clip scoring, face-tracked crop to 9:16 / 16:9 / 1:1, Remotion captions, and an MCP server with 26 tools. No watermark, no minute cap, no upload. See the open-source OpusClip alternative page for side-by-side decisions. Hosted clippers charge a subscription and require you to upload the source video.

Does Podcli upload my podcast files anywhere?+

No. Transcription, clip picking, cropping, and export all run locally. The only network calls are to the Claude or Codex API for AI clip suggestions, and those are optional.

Can Podcli turn a podcast into TikTok and YouTube Shorts?+

Yes, that's the job. It picks the moments worth clipping, crops to 1080×1920 with face tracking, burns in captions, and hands you files ready for TikTok, Shorts, and Reels.

Is Podcli a free podcast clip generator?+

Yes. Podcli is $0 under AGPL-3.0: no watermark, no minute cap, no subscription. The only optional cost is your own Claude or OpenAI API key if you want AI clip scoring.

What does OpusClip pricing look like versus Podcli?+

OpusClip typically runs about $15–$29/mo after free-tier watermarks and minute caps. Podcli is free, local, and uncapped. See what free actually costs.

How is Podcli different from ClipsAI or SupoClip?+

ClipsAI is a Python library; SupoClip is an open-source clipper with a hosted upsell. Podcli ships three entry points (CLI, web UI, MCP server) plus a knowledge base, so Claude or Codex picks clips that match your show's voice and rules, not transcript heuristics alone.

Is Podcli a podcast CLI I can run from the terminal?+

Yes. Podcli is a command-line tool first: one command (podcli process episode.mp4 --top 5) transcribes, scores, crops, captions, and exports. The web UI and MCP server sit on top of the same CLI, so you can script it, cron it, or pipe it into your own build.

What platforms does Podcli run on?+

macOS, Linux, and Windows. Hardware encoding uses VideoToolbox on Mac, NVENC on NVIDIA, and VAAPI on Linux, with a CPU fallback for everything else.

What is the MCP server for?+

Podcli ships as a Model Context Protocol server with 26 tools. Plug it into Claude Code, Codex, Cursor, or any MCP client and tell the agent to clip an episode. It handles transcription, scoring, and export end to end.

Is Podcli free and open source?+

Yes. AGPL-3.0, source on GitHub. A commercial license is available if you need to use Podcli without AGPL terms. Email support@podcli.com.

Start in your browserThree ways to use it

A simple studio in your browser, or the CLI and AI agents if you'd rather.CLI, Web UI, or AI agent. Same pipeline.

Install
curl -fsSL https://podcli.com/install.sh | sh
No prerequisites, provisions everything on first run. macOS (Apple Silicon), Linux (x64 / arm64), Windows (x64). Optional: Claude Code or Codex for AI clip suggestion.
1
Set video Drag-drop or enter a local file path
2
Add transcript Paste text, drag .txt/.srt/.vtt, or auto-transcribe
3
Find moments Claude or Codex returns scored clips from your KB
4
Review Toggle clips, pick caption style, crop mode, logo
5
Export Render with HW accel, generate titles + descriptions
Speaker detection, asset management, and knowledge base editor built into the UI header.
Process an episode
Common flags
--top 8How many clips to make
--caption-style brandedbranded · hormozi · karaoke · subtle
--crop faceSpeaker-aware face tracking, or center
--quality highlow · medium · high · max
--output ./shortsWhere clips are written
--transcript file.txtUse an existing transcript
--preset myshowLoad a saved config
--logo name | pathBrand logo to bake in
Manage
podcli assets add mylogo logo.pngRegister a logo by name
podcli presets save myshowSave the current flags
podcli info Show config and hardware
Built with
TypeScript
Python
FFmpeg
Whisper
OpenCV
Remotion
React
Express
MCP SDK