/ce-explain

/ce-explain — “Turn concepts, diffs, or recent work into dense visual explainers with optional verification.” Produces a dense, visual explainer (artifact-explainer) of a concept, a diff, or recent work, optionally verifying its claims against the code.

It implements stage-learn as the human-facing cousin of ce-compound: where compound writes machine-consumable learnings for future agents, explain writes human-consumable learnings for the developer (and teammates). Both harvest completed work into durable knowledge that improves the next iteration — the two directions of pattern-knowledge-compounding (agent grounding vs personal/team understanding). No cross-framework counterpart is paged; MP’s mp-handoff is the nearest relative (compact knowledge for a future reader), but handoff targets session continuity, not learning.

See Also

• ce-compound — the agent-facing learning sibling; mp-handoff — the compact-for-a-future-reader relative.
• artifact-explainer — the output; pattern-knowledge-compounding — the technique.
• stage-learn — the canonical stage this implements.