{"version":"2.1.235","anchor":"artifact-vm-host-bridge-sanitizes-inputs-inside-the-vm-realm","canonical_anchor":"artifact-vm-host-bridge-sanitizes-inputs-inside-the-vm-realm","heading":"Artifact VM host bridge sanitizes inputs inside the VM realm","tier":"internal","area":"Elsewhere","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.235\/e\/artifact-vm-host-bridge-sanitizes-inputs-inside-the-vm-realm","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.235","markdown":"### Artifact VM host bridge sanitizes inputs inside the VM realm\n\nArtifact sandbox host-call bridge now sanitizes inputs inside the VM before host code ever touches them\n\nThe host-call wrapper generated for the artifact sandbox now takes an optional sanitize function that runs on the VM side before the host function sees the input, so host-realm code, such as zod parsing and the consent parameter walk, no longer reads through live VM proxy traps directly. Sanitize failures, whether from length caps or unreadable traps, take the wrapper's normal error path rather than a special one. A long in-source comment marks this as a security-hardening change.\n\n- Area: Elsewhere\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5"}