mnemo-cma 0.4.3

Anthropic CMA-Memory drop-in compatibility shim — mount, mirror, export the Markdown-tree filesystem-of-memory beta announced 2026-04-23 (v0.4.1 P0-2).
Documentation

v0.4.1 (P0-2) — Anthropic CMA-Memory compat shim.

Anthropic shipped Context-Managed Agent (CMA) Memory in public beta on 2026-04-23 with anthropic-sdk-python 0.97.0 + claude-agent-sdk 0.1.68. The data model is a Markdown filesystem tree at <root>/.memory/ with a sibling audit.jsonl log. The beta is the most direct competitive surface mnemo has seen all month — operators who pick CMA for Anthropic's branding can keep using the SDK while mnemo provides:

  1. The Markdown tree on disk (read-through / write-through / mirror modes — see [SyncMode]).
  2. A bridged audit log: every CMA write produces exactly one mnemo AuditEvent whose prev_hash chains into the engine's HMAC ledger.
  3. A one-shot importer that walks an existing CMA tree, ingests every Markdown file as a Mnemo memory, and stamps the bridge head into the audit log.
  4. An export back to a byte-identical CMA tree so users can leave mnemo cleanly.

Tested against claude-agent-sdk-python==0.1.68 (pinned dev-dep in the repo's Python SDK + a "tested-against" badge in the README — drift-watcher CI is tracked under issue #cma-drift).