Skip to main content

Shared reasoning memory for parallel multi-agent systems

Project description

Mackin (Python)

Shared reasoning memory for parallel multi-agent systems.

Install

pip install mackin

For enhanced embeddings with Anthropic API:

pip install mackin[anthropic]

Quick start

from mackin import MackinStore

memory = MackinStore(session_id="my-task")

# Agent A writes reasoning
await memory.write(
    agent="researcher",
    decision="use Q4 2024 earnings report",
    reasoning="Most recent and authoritative source",
    confidence=0.9,
    tags=["source-selection"]
)

# Agent B reads structured context
context = await memory.read(
    agent="writer",
    filter={"tags": ["source-selection"], "confidence_min": 0.8}
)

# Detect conflicts
conflicts = await memory.conflicts(topic="revenue")

# Full trace
trace = await memory.trace()

No API key required

Mackin works out of the box with local hash-based embeddings. Set ANTHROPIC_API_KEY for higher-quality semantic search via Claude embeddings.

Full documentation

See the main repository README.

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

mackin-0.1.0.tar.gz (7.2 kB view details)

Uploaded Source

Built Distribution

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

mackin-0.1.0-py3-none-any.whl (8.1 kB view details)

Uploaded Python 3

File details

Details for the file mackin-0.1.0.tar.gz.

File metadata

  • Download URL: mackin-0.1.0.tar.gz
  • Upload date:
  • Size: 7.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.10

File hashes

Hashes for mackin-0.1.0.tar.gz
Algorithm Hash digest
SHA256 c0723f5820795fdb587faaaefdca735ab68244371fdc39acae4ed19adc1e2470
MD5 5a1f69d83b1cc5144363efbe5501eed2
BLAKE2b-256 32b93651048dee50b0a6ff47d55a8b4869f398aaff565a7150d64a7039bfa428

See more details on using hashes here.

File details

Details for the file mackin-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: mackin-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 8.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.10

File hashes

Hashes for mackin-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 778c25dbac1c6c69cf3439ba3fd72d345fd58db11b3eca1ef5dabc384f2e7ebf
MD5 a6a62b420792949f7b8fc08ed7bd27bf
BLAKE2b-256 157871505838a261e93d4078f8de1a3e946920ef17a5e383cfb66510417574e3

See more details on using hashes here.

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