AIMovieToolkit

How AI video tools plug into a pipeline

The pipeline register as a data file

All 255 readings, nine hooks for 28 tools, sit in one CSV. Every row names the documentation page behind it and the reading day, and the file carries a CC BY 4.0 licence. As of 2026-09-25.

What a row recordsEach row ties one tool to one hook, with the documentation page and the part of it that was read.RowIdTool, hook, counter.ReadingThe hook, in a sentence.SourcePage, part, day read.
Fig. 1 Tool, hook and question pages are all built from these rows.
Columns in pipeline-hooks.csv.
ColumnHolds
fact_idTool, hook and a counter
valueThe reading in a sentence
applies_toThe part of the page it comes from
source_urlThe documentation page
source_nameMaker and page
value_sinceThe day read
verifiedThe day read

Inclusion rule. Every column in the file. Order. File order.

1Reading the rows

An id such as luma-handoff is Luma's reading on hand-off to editing software; a trailing number marks a second reading on the same hook. The applies-to column names the part of the page the sentence comes from.

Hooks a tool's pages do not document are rows too, with an empty value and the register's nothing-read wording, so gaps can be counted.

Quotes are copied as they stand, including code fragments and endpoint names where that is how the documentation states the hook.

Documentation for developer platforms changes more often than marketing pages. Treat a row as a pointer to the page, and check the page before relying on it.

Rows with empty values are kept on purpose. A column of gaps for one hook shows at a glance how rarely that part of a pipeline is documented.

Five tools have only one or two readings. That reflects how little their introductory pages say about pipelines, not how little the platforms can do; their reference documentation may say more.

Filtering on the hook part of the id gives, for instance, every resolution reading in one list, from 1080p to 4K.

Download pipeline-hooks.csv

2Sections

3Where a drama generator fits

Read against “Data”, this is how SceneMixer describes the matching stage of its own pipeline:

Scene order, dialogue and beats follow the script; what the pipeline adds is coverage — which shot sizes, who is in frame, how long each shot holds.

SceneMixer, Script to Video, End to End — “How faithful is the shot list to my script?” (read 2026-10-13)

You can recomposite with different end-card text, and in step-by-step mode you can reshoot individual segments. The finished file can also be downloaded and shared with a link.

SceneMixer, How to Make an Ad from One Photo — “Can I change the ad after it is finished?” (read 2026-10-13)

The quote column. Each reading in the file also carries the words on its source page that it was read from, so any row can be checked against that page. Rows recorded as not stated on the pages read leave it empty.