Reelary

io.github.reelaryPublished by GitHub account reelaryv0.1.1Updated Oct 25, 2026

Search your own videos and photos by meaning and cut them into a video, 100% local.

VerifiedSTDIODesktop onlyFiles & StorageKnowledge & MemoryMedia & Design

Overview

AI-generated overview

Reelary indexes your own local videos and photos so an assistant can search them by meaning, build rough cuts from a script, and export videos.

What you can do
Check app status, plan, indexed folders and library size, and poll background jobs for indexing or export progress. List, add and remove folders in the local media library, with indexing running in the background. Search your own footage and photos by meaning, returning clips with exact start and end seconds. Turn a script into a rough cut with one scene per line, each with a best clip and two alternatives, then save it as a project. Open the Reelary window to hand-edit a project. Export a saved project or draft as an MP4, or as FCPXML or EDL for Premiere Pro, Final Cut Pro or DaVinci Resolve. List recognised people and the clips each appears in, and list places where footage was shot.
When to use it
Useful when you edit video from your own footage and want an assistant to find relevant clips by description, assemble a rough cut from a script, or produce a recap or vertical short. Also suited to exporting an edit into a desktop NLE.
Requirements
Runs as a local process on a Mac or Windows desktop via npx reelary, which installs the signed Reelary app that runs in the background. No API keys or environment variables are declared. A Reelary plan applies: free covers unlimited indexing, visual search and rough cuts; Pro unlocks dialogue search, people, places, NLE export and full-quality unlimited exports.
Before you install
Adding a folder indexes its contents into the local library; removing a folder only drops it from the index and leaves files on disk. Exports write MP4 or NLE files, and saving creates projects in the app. Free-plan MP4 exports are limited to 720p, a watermark, 60 seconds and 5 per month; Pro is a one-time purchase. The README states processing is 100% local and nothing leaves the computer.

Installation

In SourceWeft

  1. Open Reelary in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Reelary for AI agents

Talk to your footage. Reelary indexes the videos and photos already on your Mac or Windows PC and lets Claude, Cursor, Codex and other agents search every frame by meaning and cut the matches into a video. 100% local: nothing leaves your computer.

Install

npx reelary

Installs the signed Reelary app (runs in the background with a tray icon), connects it to Claude Code, Claude Desktop, Cursor and Codex, and adds the skills below.

Skills only (any of 60+ agents):

npx skills add reelary/skills

Claude Code plugin (skills + MCP server):

/plugin marketplace add reelary/skills/plugin install reelary@reelary

What you can ask

  • "Index ~/Movies and find drone shots of the coast."
  • "Find b-roll for this voiceover: …"
  • "Cut a 30-second 9:16 recap of my Lisbon trip."
  • "Rough cut this script from my footage and export it for DaVinci Resolve."

Skills

SkillFor
reelaryStart here: install, index folders and drives, troubleshoot
find-footageShots by description, photos, quotes (Pro), people (Pro), places and drone GPS (Pro)
broll-finderB-roll for a voiceover or interview
rough-cutRough cuts, script-to-video, montages, highlight reels
vertical-videoShorts, Reels, TikTok (9:16, 1:1)
nle-exportFCPXML / XML / EDL for Premiere, Final Cut, Resolve (Pro)

Skills source: https://github.com/reelary/skills

Plans

Free: unlimited indexing, visual search, rough cuts; MP4 exports at 720p with a watermark, up to 60 seconds, 5 per month. Pro (one-time purchase, Mac + Windows): dialogue search, people, places, NLE export, full-quality unlimited exports. https://reelary.app/pricing

CLI

npx reelary statusnpx reelary index ~/Moviesnpx reelary search "sunset over water" --jsonnpx reelary cut "Sunrise over the city. Coffee being poured." --aspect 9:16 --save Morningnpx reelary export <project_id> [--nle fcpxml]npx reelary uninstall [--app]

Source: README.md at commit 27bb61f

Tools

14

reelary_status

Reelary app status: version, free/pro plan, indexed folders and library size, running index/export jobs, export limits. Call first if unsure Reelary is ready.

Takes no parameters.

reelary_list_folders

List the folders in the user's local Reelary media library.

Takes no parameters.

reelary_add_folder

Add a folder of videos/photos to the local library and index it (runs in the background; poll reelary_job_status). Indexing is free and unlimited.

Parameterspathrequired

reelary_remove_folder

Remove a folder from the library index (files on disk are not touched).

Parameterspathrequired

reelary_rough_cut

Turn a script into a rough cut from the user's own footage: one scene per line, each with the best clip and 2 alternatives. Returns a draft to save or export.

Parametersaspectscriptrequiredmax_clip_seconds

reelary_save_project

Save a rough-cut draft as a Reelary project the user can open and fine-tune in the editor.

Parametersnamerequireddraftrequired

reelary_open_in_app

Show the Reelary window, optionally opening a saved project in the editor for hand edits.

Parametersproject_id

reelary_export_video

Render an MP4 from a saved project (preferred, keeps the user's edits) or a draft. Runs in the background; poll reelary_job_status. Free plan: 720p, watermark, 60s, 5 per month.

Parametersdraftaspectout_pathoverwriteproject_id

reelary_export_nle

Export a project or draft as an editing project for Premiere Pro / Final Cut Pro / DaVinci Resolve (FCPXML or EDL), referencing the original files. Pro.

Parametersdraftformatout_pathoverwriteproject_id

reelary_job_status

Progress of a background job (indexing or export).

Parametersjob_idrequired

reelary_list_people

People recognised in the library (face search). Pro.

Takes no parameters.

reelary_person_clips

Every clip a recognised person appears in. Pro.

Parametersperson_idrequired

reelary_list_places

Where footage was shot (GPS from phones, drones, GoPro) with place names. Pro.

Takes no parameters.

Version history

1
  1. v0.1.1LatestOct 24, 2026

Score

80% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts
  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.