ArtAgents is a file-based toolkit for producing Reigh-compatible video edits, event-talk renders, generative timelines, and image/video assets.
The core idea is simple: conductors orchestrate, players execute. You ask an agent for an outcome, and ArtAgents routes the work through the right existing nodes or newly built nodes.
Copy this into a coding agent:
Use the ArtAgents repo:
https://github.com/banodoco/ArtAgents
First read README.md and SKILL.md, then run:
git status --short
Keep generated files under runs/.
Do not commit secrets or media.
Run:
python3 pipeline.py --brief brief.txt --out runs/example --render
While working, call out friction points, suggest fixes, and recommend PRs to
the original upstream repos when the durable fix belongs there rather than as
a local workaround.
# Source-video hype cut
python3 pipeline.py --video source.mp4 --brief brief.txt --out runs/example --render
# Audio-backed timeline
python3 pipeline.py --audio rant.wav --brief brief.txt --out runs/audio --render
# Pure-generative timeline
python3 pipeline.py --brief brief.txt --theme 2rp --out runs/generative --render --target-duration 28
# Event-talk render
python3 bin/event_talks.py render --manifest runs/event/talks.json --out-dir runs/event/renderedGenerated outputs belong under runs/. That directory is ignored by git and can
contain frames, JSON files, audits, and rendered videos from local experiments.
# Rerun a pipeline from a specific stage
python3 pipeline.py --video source.mp4 --brief brief.txt --out runs/example --from cut --render
# Audit a run
python3 pipeline.py audit --run runs/example
python3 pipeline.py audit --run runs/example --json
# Inspect available conductors and performers
python3 pipeline.py conductors list
python3 pipeline.py performers list
# Fetch canonical Reigh data through the app Edge Function
python3 pipeline.py reigh-data --project-id <PROJECT_UUID> --shot-id <SHOT_UUID> --out runs/reigh/shot.json
# Render a Reigh-compatible timeline/assets pair
python3 bin/render_remotion.py \
--timeline runs/example/briefs/my-brief/hype.timeline.json \
--assets runs/example/briefs/my-brief/hype.assets.json \
--out runs/example/briefs/my-brief/render.mp4artagents/ Python implementation
artagents/conductors/ Workflow orchestration
artagents/performers/ Executable player actions
bin/ Direct stage launchers
remotion/ Remotion renderer project
scripts/ Development and code-generation scripts
examples/ Small schema fixtures
_reference/ Copied Reigh contract references
runs/ Ignored local outputs
pipeline.py is the primary entry point. bin/*.py launchers call the matching
modules under artagents/ when you need a single stage directly.
The main pipeline writes source-level artifacts into the run directory and
brief-level artifacts under briefs/<slug>/:
runs/example/
transcript.json
scenes.json
shots.json
pool.json
audit/
briefs/
my-brief/
arrangement.json
hype.timeline.json
hype.assets.json
hype.metadata.json
hype.mp4
validation.json
python3 -m pip install -r requirements.txt
pytest
cd remotion
npm run typecheck
npm run smoke
npm run gen-typesRun npm run gen-types after effect, animation, or theme primitive changes.
Local secrets belong in .env, .env.*, or this.env; these are ignored by
git. Large source media and generated artifacts should stay under runs/ or
another ignored output directory.
Open Source Native License (OSNL) v0.2. See LICENSE for the full terms.
