Summary
Every scan fails at container startup. The CLI reports Scan failed — exec transport error; the sandbox container exits with code 1. Root cause is inside the sandbox image: the caido proxy generates a CA certificate, then fails to re-import it with Certificate error: InvalidP12 ("header too long / bad object header"), and the process aborts.
The host toolchain is fully healthy — I verified DNS, filesystem read/write, and OpenSSL p12 generation/parsing all work correctly inside the same image. The failure is internal to caido's cert handling in the image.
Environment
- Strix CLI: 1.5.3 (installed via
curl -sSL https://strix.ai/install | bash)
- Sandbox image: ghcr.io/usestrix/strix-sandbox:1.3.0 (
latest resolves to the same digest sha256:f6906c31…)
- Host: macOS 13 (Ventura, Darwin 22.6.0), Intel x86_64
- Docker runtime: colima (macOS Virtualization.Framework / vz), Docker server 29.5.2, API 1.54
DOCKER_HOST=unix://$HOME/.colima/default/docker.sock
Steps to reproduce
strix --target ~/some-local-repo --scan-mode quick
Summary
Every scan fails at container startup. The CLI reports
Scan failed — exec transport error; the sandbox container exits with code 1. Root cause is inside the sandbox image: the caido proxy generates a CA certificate, then fails to re-import it withCertificate error: InvalidP12("header too long / bad object header"), and the process aborts.The host toolchain is fully healthy — I verified DNS, filesystem read/write, and OpenSSL p12 generation/parsing all work correctly inside the same image. The failure is internal to caido's cert handling in the image.
Environment
curl -sSL https://strix.ai/install | bash)latestresolves to the same digestsha256:f6906c31…)DOCKER_HOST=unix://$HOME/.colima/default/docker.sockSteps to reproduce
strix --target ~/some-local-repo --scan-mode quick