section trustlanguage plainunknowns explicit
Trust begins with a clear boundary
Clear boundaries matter more than broad promises. Smith Labs states what a service accepts, what it records, what it cannot conclude, and where a human decision is still required.
Operating principles
Minimum necessary information
This website has no analytics, advertising, application trackers, or cookies. Hosting and security systems still process ordinary request metadata.
Evidence before certainty
SmithGuard preserves source, timing, confidence, corrections, and explicit unknowns. Early evidence is visibly labeled Emerging / Unverified.
Read-only public boundary
The public SmithGuard service answers questions about public artifacts. It does not accept private code or authorize changes to another system.
Human decisions remain human
Documentation, records, and recommendations provide information. They do not grant an agent or person permission to take a high-risk action.
Corrections remain visible
A changed conclusion creates a new revision. Earlier evidence is retained so users can understand what changed and why.
Report a security concern
Email contact@smithlabs.ai. Include enough detail to reproduce or assess the concern, but do not send credentials, private keys, personal data, or destructive proof.