recall

To retrieve past session context, use temporal drill-down to navigate from year to day layers. This skill reconstructs interaction history from temporally indexed summaries, loading Markdown bodies as instructions to revive details not stored as files.

Drilling down through temporal layers is done by navigating across year, quarter, month, week, and day intervals. This multi-hop navigation allows you to progressively narrow down and locate the exact session content or commands you need.

Timeline search works by enforcing discovery via YAML frontmatter containing a name and description, then supporting multi-hop navigation across temporal layers. It loads the bottom Markdown body as operational instructions to reproduce exact session content.

Yes, you can answer questions using persistent recall based on first-person experience and cached context. By drilling down to the previous week's temporal layer, you can surface key details and reproduce exact commands from past interactions.

Yes, YAML frontmatter with a name and description is required for discovery. The temporal drill-down relies on this frontmatter to index sessions and load the corresponding Markdown body as operational instructions for context reconstruction.

The main limitation is that recall depends on temporally indexed summaries and YAML frontmatter discovery. If past sessions lack these indexed layers or frontmatter metadata, the multi-hop navigation cannot reconstruct or revive the lost context.