> ## Documentation Index > Fetch the complete documentation index at: https://site.aspect.build/llms.txt > Use this file to discover all available pages before exploring further. # Selective Delivery for Bazel > Deliver only the targets whose Bazel-built outputs actually changed. Digest-based change detection, correctly stamped artifacts, parallel pushes, and a delivery manifest. export const MarketingPage = () =>
; export const CTA = ({title, subtitle, primaryCta, primaryHref = "#", secondaryCta, secondaryHref = "#"}) =>

{title}

{subtitle &&
{subtitle}
}
{primaryCta && {primaryCta} } {secondaryCta && {secondaryCta} }
; export const FeatureCard = ({title, description, href, children}) => { const Tag = href ? "a" : "div"; const tagProps = href ? { href } : {}; return {children &&
{children}
}

{title}

{description &&

{description}

} ; }; export const FeatureGrid2 = ({children}) =>
{children}
; export const FeatureGrid3 = ({children}) =>
{children}
; export const SectionHeader = ({title, subtitle, centered = true, label}) =>
{label &&

{label}

}

{title}

{subtitle &&
{subtitle}
}
; export const Section = ({children, className = "", gray = false, dark = false, id}) =>
{children}
; export const Hero = ({title, subtitle, badge, primaryCta, primaryHref = "#", secondaryCta, secondaryHref = "#", centered = true, children}) =>
{centered && !children ?
{badge &&
{badge}
}

{String(title).split(/\\n|\n/).map((line, i) => i ? [
, line] : line)}

{subtitle &&
{subtitle}
}
{primaryCta && {primaryCta} } {secondaryCta && {secondaryCta} }
:
{badge &&
{badge}
}

{String(title).split(/\\n|\n/).map((line, i) => i ? [
, line] : line)}

{subtitle &&
{subtitle}
}
{primaryCta && {primaryCta} } {secondaryCta && {secondaryCta} }
{children &&
{children}
}
}
;

Pushing every artifact on every green build is slow and risky. You overwrite known-good artifacts with identical but unverified rebuilds, and you burn CI time, registry storage, and network egress on uploads that change nothing.

Bazel already knows exactly what changed. Selective Delivery applies that knowledge at release time, comparing output digests against the prior release and pushing only the difference.