-
Constraint-Satisfaction Enthusiasts who want their SSG to "solve" for the best layout rather than just render it.
-
Researchers exploring the intersection of lazy evaluation, non-deterministic search, and logical variables.
-
Architects needing a generator capable of handling ambiguous content relationships and complex taxonomies.
-
Teams seeking a "search-first" approach to static content orchestration.
Define your site structure as a set of constraints (e.g., "This post must be adjacent to a related image, and the total page-weight must not exceed 200kb"). The engine "searches" the solution space to generate the site automatically.
Leveraging the Curry language, you get the best of both worlds: the strictness of algebraic data types and the power of logical variables. This allows for complex, high-level site queries that are impossible in traditional imperative SSGs.
If a layout has multiple valid configurations, vindaloo-ssg can explore all possibilities (non-determinism). This is perfect for A/B testing or generating multiple variations of a site from a single source-logic core.
# Setup the Curry environment (PAKCS or KiCS2) via asdf
just setup
# Run the constraint-based build
# The engine will 'solve' for the optimal site structure
just build
# Type-check and verify constraints without building
just check-
Logic-Driven Routing - Use logical variables to define dynamic site paths and relationships.
-
Lazy Evaluation - Only render what is strictly necessary to satisfy the current search constraints.
-
Podman-First - Isolated, containerised logic-search environments for Linux, Minix, and PC.
-
Accessibility - Integrated support for BSL, GSL, ASL, and Makaton metadata.
-
Edge-Ready - Verified for ASICs and Edge tech via efficient backend emission.
vindaloo-ssg provides the constraint-logic and non-deterministic tier of the poly-ssg family.