A web audio tool built for both humans and AI agents
AwsmAudio is a web tool for making and editing sound. A person can use its visual , while an AI agent can control the same sound system through MCP.
It can handle simple requests like making a retro-style jazz track, but the stronger use case is sound effects for games or . For example, it could create a jetpack sound that changes while the game runs, based on a character’s health or remaining fuel.
The target user is less a traditional musician and more a who wants to shape sound with an AI agent, then control that sound from code. The tool maps WebAudio building blocks into both a human and an agent-facing control layer.
Key points
- AwsmAudio is a sound and editing tool.
- Humans use the , while an AI agent can control the tool through MCP.
- The main use case is dynamic sound effects for games and , not general music .
- Sounds can change based on live game data, such as health or fuel level.
- The approach gives AI agents their own control layer instead of only adding chat on top of an existing tool.