list_films
list films in workspace
LABSCO SUMMARY
The consistency loop is the part nobody else has: generate, judge, fix, promote.
AI video's real failure is not quality, it is that the character's jacket changes between shots. Reel Studio makes that a first-class workflow — takes are reviewed against consistency verdicts, the failing ones produce a fix prompt automatically, and you promote the take that passed. Combined with named cast and location bindings written into the shot prompt, that is a genuine answer to continuity rather than a hope that the seed holds. Check credits with get_workspace_status before a batch; generation jobs are where the money goes.
What it is
Reel Studio's endpoint covers AI filmmaking end to end rather than exposing a generate call. Films hold scenes, scenes hold shots, and cast and locations are named entities that shots bind to.
What you get
- Structure — films, scenes, shots, cast and locations, each with list, create and update tools
- Binding — a shot description references cast and locations by name, so the same character stays the same character
- Generation — a model catalogue spanning image, text-to-video, image-to-video, video-to-video, avatar, lipsync, speech, sound effects and music, with jobs you poll
- Consistency — review_takes reads the model's verdicts across a shot's history and recommends the best passing take; promote_take makes it active; generate_fix_prompt turns a flagged issue into a fix prompt
- Timeline — add, edit, reorder and move visual clips, plus voiceover, effects and music cues with per-lane mute and volume
- Export — start a render and poll for the final video URL
Requirements
OAuth 2.1 with PKCE, or a personal access token. get_workspace_status reports the authenticated user, the selected workspace and available credits before you spend any. Work is private by default; custom access controls are handled by email.
Setup effort
Paste a URL, then authorize — add the endpoint to your client, then approve the OAuth consent screen
USE CASES
01
Building the film's world
Cast and locations as named entities that shots can bind to.
02
Writing the shotlist
Scenes seed shotlists; shot descriptions reference cast and locations by name.
03
Generating and judging
Kick off a job, poll it, then read the consistency verdicts across takes.
04
Fixing a bad take
Turn the flagged issue into a fix prompt, then promote the take that passes.
05
Cutting and shipping
Assemble the visual track, place audio cues, set the mix, render.
40 tools
Listed in project docs. Your client may show a different list.
get_film
get a film by slug
create_film
create a new film
get_workspace_status
confirm authenticated user, selected workspace, accessible workspaces, and available credits before spending
list_scenes
list scenes for a film
create_scene
scene authoring; outline is the scene body/prompt used to seed shotlists and scene-frame generation
update_scene
update a scene
list_shots
list shots for a film
create_shot
shot authoring; description is the shot prompt seed; use @{Cast Name} and #{Location Name} bindings
update_shot
update a shot
list_cast
list cast members
create_cast
character authoring
update_cast
update a cast member
list_locations
list locations
create_location
location authoring
update_location
update a location
list_assets
list uploaded image/video/audio assets for a film
create_asset_upload
step 1 of local asset upload: create a presigned R2 upload URL
confirm_asset_upload
step 3: confirm asset after direct-to-R2 PUT succeeds
upload_asset
import an existing HTTP(S) media URL into the film asset library
list_models
model catalog (image/t2v/i2v/v2v/avatar/lipsync/tts/sfx/music) with allowed aspect ratios, durations, resolutions, per-model credit costs
generate
kick off a generation job; targets include film-poster, shot-frame, shot-video, scene-frame, scene-video, cast-portrait, cast-sheet, location-image, location-sheet, timeline-sfx, timeline-voiceover, film-score, standalone; up to 4 parallel takes
get_job
poll job status
list_jobs
list jobs
review_takes
read Reel-VLM consistency verdicts from a shot's frame/video history and recommend best passing take
promote_take
set a reviewed take as active frameUrl/videoUrl
generate_fix_prompt
turn a VLM consistency issue into a Frame Lab fix prompt (action: edit|regenerate)
get_timeline
read current V1 visual cut plus VO/A2/M1 audio cues
add_timeline_clip
add a V1 clip
update_timeline_clip
edit a V1 clip
remove_timeline_clip
remove a V1 clip
reorder_timeline
reorder V1 clips
move_timeline_clips
move V1 clips
assemble_timeline
arrange ordered shot/scene ids into a cut; can replace existing V1 track
add_timeline_audio_clip
add an existing/generated audio URL as VO/A2/M1 cue
update_timeline_audio_clip
edit VO/A2/M1 audio cue timing/trim/gain
remove_timeline_audio_clip
delete an audio cue
update_audio_mix
set lane-level mute/volume for A1 dialogue, VO, A2 SFX, M1 music
start_export
render current cut
get_export
poll export job until final video URL ready
BEFORE YOU INSTALL
What to know first
Tools you get
40 tools named in the project's documentation.
Requires
- an MCP client that can reach a remote endpoint
- OAuth 2.1 with PKCE and dynamic client registration, or a personal access token as a bearer header
INSTALLATION
Connect your client
Add the endpoint to your client, then approve the OAuth consent screen.
Server setup
ENDPOINT URL
https://reel.studio/api/mcpRequirements
- OAuth 2.1 with PKCE and dynamic client registration, or a personal access token as a bearer header
