agentsclimarketplace

Bloom filter visualization pattern

Skill kjuhwa/skills-hub/skills/design/bloom-filter-visualization-pattern

Self-correcting knowledge corpus for Claude Code — 9 stable shape clusters, bias-correction pipeline baked into contribution flow. 47 papers, 45 techniques, 1.1k skills.

Install
npx -y skills add kjuhwa/skills-hub --skill bloom-filter-visualization-pattern

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Visualizing bloom filter bit arrays, hash function mappings, and false-positive regions for interactive exploration

SKILL.md

1.9 KB, 338 tokens by cl100k_base, as published. Nobody here has run it

bloom-filter-visualization-pattern

When building an interactive bloom filter explorer, render the underlying bit array as a grid of cells (typically 64-256 bits visible) where each cell reflects its 0/1 state with distinct colors — empty cells neutral, set cells filled, and newly-flipped cells highlighted briefly on insertion. Overlay hash function outputs as colored connector lines or arrows from the input token down to the k bit positions it maps to, using one color per hash function (h1, h2, h3) so users can trace which hash caused which bit flip. This makes the "multiple hashes → multiple bits" invariant visually self-evident rather than abstract.

For query operations, animate the lookup sequence: highlight the k target bits in order, and resolve to either "definitely not present" (any bit is 0, shown in red) or "possibly present" (all bits are 1, shown in amber — never green, since false positives are possible). Surface a live fill-ratio gauge and a computed false-positive probability (1 - e^(-kn/m))^k updated after every insertion so users watch the FPR curve climb as the filter saturates. Collision/race variants should additionally color bits that are "shared" by multiple inserted items differently, making the source of false positives tangible.

Keep the visualization deterministic and replay-friendly: expose m (bit array size), k (hash count), and the hash seeds as URL-encoded parameters, and provide step-forward/step-back controls over the operation log. This lets users pause mid-insertion to inspect which hashes are about to fire, which is far more instructive than instant commits.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.