Overview
Context Passing is Spine’s core infrastructure that allows blocks to automatically share information with each other. When blocks get connected, Spine figures out the best way to pass context by preserving maximum relevant context and translating between block types seamlessly. Under the hood there is a full dependency graph that determines execution order, parallelism, and concurrency across connected blocks. Context propagation handles long branches of connected blocks and the wide variance in context sizes across different block types. At scale, this infrastructure is what keeps the workload coherent and consistent across hundreds of simultaneous operations.How Context Flows
When blocks are connected (via drag-and-drop lines), they automatically share context. Any action you take with a block will be grounded in the connected context. You can connect any block to any other block, and that connection guarantees the passing of context regardless of block type. Within the current block, you can also pass context explicitly in specific variables. This is useful when you want to tell the block exactly what to do with the connected content. For example: “Update the image [image_block_name] with the new design guidelines [design_guidelines]“Creating Connections
- Hover over a block — Connection points appear on block edges
- Click and drag — From one block’s connection point to another
Connected Block Cost
Connected blocks have a cost multiplier: for every 3 connected blocks, cost increases by 1x block price. This reflects the additional context processing required.Best Practices
- Connect related blocks — Link blocks that should inform each other
- Avoid over-connection — Too many connections can create noise
- Use strategic context — Connect blocks where information flow adds value
- Review context flow — Ensure connections are helping, not hindering