Skip to main content

Vault-native memory engine for Claude Code. FTS5 BM25, RRF-ready retrieval, gated Graphiti KG, and adaptive context tooling.

Project description

mneme-core

Python core for mneme: vault-native memory for Claude Code, also usable from Codex and any MCP client. mneme is Claude-Code-native by origin and client-neutral at the core.

This package provides:

  • FTS5 BM25 full-text indexer with language-aware normalization (Turkish casefold module included as utility).
  • RRF-ready retrieval pipeline at k=60, with FTS5 shipped and optional backends injectable by callers.
  • Graphiti bi-temporal knowledge graph adapter, gated by the full profile and local Neo4j.
  • Roadmap dense embedding adapter. The v1.0 standard profile reserves the runtime slot but does not ship packaged LEANN retrieval.
  • Background compression pipeline with 4-D rubric (Accuracy, Depth, Context, Continuity) and cost cap ledger.
  • Adaptive Context Layer: distill.shell_compress, distill.injection_dedup, distill.adaptive_topk, distill.compressed_format, and the mneme audit CLI.

Installation

pip install mneme-core

For development:

pip install -e ".[dev]"

Quality Gates

This package targets Python 3.11+ and passes ruff lint plus mypy --strict. Test coverage minimum is 80 percent for business logic.

License

MIT. See LICENSE in the repository root.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

mneme_core-3.1.0.tar.gz (282.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

mneme_core-3.1.0-py3-none-any.whl (197.0 kB view details)

Uploaded Python 3

File details

Details for the file mneme_core-3.1.0.tar.gz.

File metadata

  • Download URL: mneme_core-3.1.0.tar.gz
  • Upload date:
  • Size: 282.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for mneme_core-3.1.0.tar.gz
Algorithm Hash digest
SHA256 f2b44af8edf18640be5a450f3341bf01bfe40547a4626e19a347b43a4bc50051
MD5 700e96af17cbb8e362cbc31afdb27dba
BLAKE2b-256 b431b1d20d0672668478acfaed5bab5f54d93179362b91c0655e3f3bd3fb8d39

See more details on using hashes here.

Provenance

The following attestation bundles were made for mneme_core-3.1.0.tar.gz:

Publisher: release.yml on TheGoatPsy/mneme

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file mneme_core-3.1.0-py3-none-any.whl.

File metadata

  • Download URL: mneme_core-3.1.0-py3-none-any.whl
  • Upload date:
  • Size: 197.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for mneme_core-3.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 ebbba0632339a65322ae955fb4f025c10c7298f849a2f7892f38b56f5ba3d83d
MD5 0d0316777bb8a7a500ac0ecbd1f3d425
BLAKE2b-256 a488b7cbda66471caf5c393ff10d8ebe4b37bad2b21d264f6e78950751e49df3

See more details on using hashes here.

Provenance

The following attestation bundles were made for mneme_core-3.1.0-py3-none-any.whl:

Publisher: release.yml on TheGoatPsy/mneme

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page