wehub-resource-syncAudacity CLI
A stateful command-line interface for audio editing, following the same patterns as the GIMP and Blender CLIs in this repo.
Architecture
- JSON project format tracks state (tracks, clips, effects, labels, selection)
- Python stdlib (
wave,struct,math) handles WAV I/O and audio processing - Click provides the CLI framework with subcommand groups and REPL
- Effects are recorded in the project JSON and applied during export/render
Install
pip install click numpy # numpy only needed for tests
No other dependencies required. Core functionality uses only Python stdlib.
Run
# From the agent-harness/ directory:
cd /root/cli-anything/audacity/agent-harness
# One-shot commands
python3 -m cli.audacity_cli project new --name "My Podcast"
python3 -m cli.audacity_cli track add --name "Voice"
python3 -m cli.audacity_cli clip add 0 /path/to/recording.wav
python3 -m cli.audacity_cli effect add normalize --track 0
python3 -m cli.audacity_cli export render output.wav
# JSON output mode (for agent consumption)
python3 -m cli.audacity_cli --json project info
# Interactive REPL
python3 -m cli.audacity_cli repl
Run Tests
cd /root/cli-anything/audacity/agent-harness
# All tests
python3 -m pytest cli/tests/ -v
# Unit tests only (no real audio files)
python3 -m pytest cli/tests/test_core.py -v
# E2E tests (generates real WAV files)
python3 -m pytest cli/tests/test_full_e2e.py -v
Evaluation (Eval Harness)
Run a lightweight evaluation suite and generate reports:
# Default output: eval_results/<timestamp>/
python3 -m cli.audacity_cli eval
# Custom output directory
python3 -m cli.audacity_cli eval --out ./eval_out
# Compare to a baseline and fail on regression
python3 -m cli.audacity_cli eval --baseline baseline.json --fail-on-regression
# Update (write) baseline JSON
python3 -m cli.audacity_cli eval --baseline baseline.json --update-baseline
Outputs:
eval_report.jsonandeval_report.mdartifacts/directory for task outputs
Command Groups
| Group | Commands |
|---|---|
project |
new, open, save, info, settings, json |
track |
add, remove, list, set |
clip |
import, add, remove, trim, split, move, list |
effect |
list-available, info, add, remove, set, list |
selection |
set, all, none, info |
label |
add, remove, list |
media |
probe, check |
export |
presets, preset-info, render |
session |
status, undo, redo, history |
Example Workflow
# Create a podcast project
python3 -m cli.audacity_cli project new --name "Episode 1" -o project.json
# Add tracks
python3 -m cli.audacity_cli --project project.json track add --name "Host"
python3 -m cli.audacity_cli --project project.json track add --name "Guest"
python3 -m cli.audacity_cli --project project.json track add --name "Music"
# Import audio clips
python3 -m cli.audacity_cli --project project.json clip add 0 host_recording.wav
python3 -m cli.audacity_cli --project project.json clip add 1 guest_recording.wav --start 0.5
python3 -m cli.audacity_cli --project project.json clip add 2 music.wav --volume 0.3
# Apply effects
python3 -m cli.audacity_cli --project project.json effect add normalize --track 0 -p target_db=-3.0
python3 -m cli.audacity_cli --project project.json effect add compress --track 0 -p threshold=-20 -p ratio=4.0
python3 -m cli.audacity_cli --project project.json effect add fade_in --track 2 -p duration=2.0
# Add labels
python3 -m cli.audacity_cli --project project.json label add 0.0 --text "Intro"
python3 -m cli.audacity_cli --project project.json label add 30.0 -e 60.0 --text "Main discussion"
# Export
python3 -m cli.audacity_cli --project project.json export render episode1.wav --preset wav