Claude Transcripts docs GitHub
Work in progressUnder active development — not tested as ready for use. Breaking changes land without notice, stored data may need to be discarded between revisions, and there is no auth or security model. These docs describe the intended design as much as the current state.

15. Tiered architecture (Tier 1 / 2 / 3)

Date: 2026-06-18

Status

Accepted

Context

The project's ambitions span a wide range — from "retain my Claude Code transcripts on one laptop" to "a replicated, multi-user corpus that agents actively learn from." Trying to design for all of it at once risks either over-building the simple case or baking in single-machine assumptions that block the advanced case.

Decision

Organise scope into three stacking tiers, each a strict superset of the one below, each independently useful, with the rule that nothing in a higher tier may break a lower tier:

The full feature breakdown lives in tiers.md.

Consequences