Willow Inference Server
PassAudited by ClawScan on May 1, 2026.
Overview
This is a coherent instruction-only skill for using a local speech server, with user-directed setup and data-sending steps that should be reviewed before use.
Before installing, confirm that the GitHub repository is the intended Willow server source, review the utility script before running it, and configure WILLOW_BASE_URL only to a trusted server because audio and text sent there may contain private information.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Running the setup may install software or change the local environment based on code from the referenced GitHub repository.
Setup depends on an external, unpinned repository and shell scripts that are not included in the artifact set. This is user-directed and central to running the server, but the upstream code should be reviewed or pinned.
git clone https://github.com/toverainc/willow-inference-server.git cd willow-inference-server ./utils.sh install ./utils.sh gen-cert your-hostname ./utils.sh run
Verify the repository, review the utility script, and consider pinning to a trusted commit before running the setup commands.
Private recordings or text could be processed by whichever server is configured in WILLOW_BASE_URL.
The skill sends audio files and text to the configured Willow server endpoint. That is the expected function, but the content may be private and the endpoint choice matters.
curl -X POST "${WILLOW_BASE_URL}/api/asr" \
-F "audio_file=@/path/to/audio.m4a" ...
curl -X POST "${WILLOW_BASE_URL}/tts" ... -d '{"text": "Hello world", "voice": "af_sarah"}'Use a trusted local or self-managed Willow server, prefer HTTPS, and avoid sending sensitive audio or text to an endpoint you do not control.