Cdc visualization pattern
Skill kjuhwa/skills-hub/skills/design/cdc-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.
npx -y skills add kjuhwa/skills-hub --skill cdc-visualization-patternAssembled 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 CDC event flow from source database through log capture to downstream consumers with row-level change granularity
SKILL.md
1.9 KB, 335 tokens by cl100k_base, as published. Nobody here has run it
cdc-visualization-pattern
CDC (Change Data Capture) visualizations must render three distinct layers simultaneously: the source database with its transaction log (WAL/binlog/redo), the CDC connector/capture process reading log positions (LSN/GTID/SCN), and downstream sinks (Kafka topics, search indexes, data lakes). Use horizontal swim lanes where each lane represents a stage, and animate discrete change events (INSERT/UPDATE/DELETE) as colored tokens flowing left-to-right. Color code by operation type: green for INSERT, amber for UPDATE, red for DELETE, and purple for schema changes (DDL). Always surface the log position cursor prominently — it is the single most important state in any CDC system.
For cdc-stream-monitor style real-time dashboards, show lag as the visual gap between the "current log tail" marker and the "last consumed position" marker per consumer. For cdc-replay-theater style time-travel UIs, render a scrubber timeline keyed to LSN/timestamp with event density histograms so users can see hot windows before seeking. For cdc-topology-weaver style graph views, draw source tables as nodes with edges representing subscription routes to sinks; overlay edges with throughput sparklines and a "last heartbeat" age badge so dead routes are instantly visible.
Always expose the before/after row image side-by-side for UPDATE events — a CDC tool that only shows the "after" state is half-blind. For DELETEs, preserve the tombstone marker visually (struck-through row with retention-until timestamp) rather than removing it from view, because downstream replay windows depend on tombstone visibility.