v0.1.2

Release Notes for v0.1.2

This release adds automatic rule-based memory-type classification, ships two new first-class integrations (MCP server and Hermes Agents), unifies content-length and kiosk_mode defaults across REST/SDK/CLI, adds a chronological recall --recent path, and renames the CrewAI package on PyPI.

New Features

• Configurable rule-based memory parsing (memory_parsing_service.py)

  • New MemoryParsingService auto-detects a memory's type at ingestion using
    score-based classification (with priority tie-breaking) across all 13
    supported memory types — no more blind default to fact.
  • MemoryRecord.type is now optional (None); the parser assigns the type
    when the caller omits it.
  • New AUTO_PARSE_ENABLED setting (default True).
  • remember and batch-remember run the parser when type is omitted and
    now return the resolved type in the response.
  • CLI memanto remember no longer forces --type fact; it displays the
    parsed type instead.

• MCP server integration (integrations/mcp/)

  • New memanto_mcp package exposing Memanto memory operations to MCP clients,
    with config, lifecycle, server, and tools modules.
  • Includes packaging (pyproject.toml), LICENSE, .env.example, and a test
    suite (test_cli, test_config, test_server).

• Hermes Agents integration (integrations/hermes-agents/)

  • New hermes_memanto provider (provider.py) with an installer
    (install.py), plugin.yaml, packaging, LICENSE, and provider tests.

Improvements

• Unified content-length cap across layers (direct_client.py, sdk_client.py,
  config.py)

  • SDK/Direct clients now use InputLimits.MAX_TEXT_LENGTH instead of a
    hardcoded 500, aligning the cap with the REST/Pydantic models (10,000 chars).
  • Removed the unused MAX_MEMORY_SIZE / MAX_TITLE_SIZE settings.

• recall --recent chronological retrieval (memory.py, clients)

  • New recall_recent() on SdkClient and DirectClient returns the most
    recently stored memories (newest first).
  • New memanto recall --recent CLI flag — lists recent memories directly, no
    search query required (mutually exclusive with --as-of / --changed-since).

• Unified kiosk_mode + threshold defaults (memory.py, clients)

  • kiosk_mode and threshold defaults now resolve from config.yaml.
  • threshold is only applied when kiosk_mode is on; the kiosk-mode fallback
    threshold is unified to 0.15 across REST and config defaults.

• CrewAI integration rename & tooling constraints (integrations/crewai/)

  • PyPI package renamed memanto-crewai → crewai-memanto; package directory
    memanto_crewai → crewai_memanto.
  • LLM tool schemas now enumerate all 13 memory types with definitions to guide
    classification.

• Docker / packaging (Dockerfile, pyproject.toml)

  • Dockerfile accepts a VERSION build arg (--build-arg VERSION=x.y.z) feeding
    SETUPTOOLS_SCM_PRETEND_VERSION.
  • hatch-vcs version raw-options configured and integrations/ excluded from
    the main package's version discovery.

Tests

• Added tests/test_memory_parsing.py for the parsing service, plus new MCP and
  Hermes integration test suites; expanded test_api.py and test_cli.py.

Full Changelog

Full Changelog: v0.1.1...v0.1.2