Metadata-Version: 2.4
Name: vectorharness
Version: 0.1.0
Summary: Continuous time-decay semantic memory for autonomous AI agents.
Author-email: OmniAxes <support@omniaxes.online>
Project-URL: Homepage, https://vectorharness.omniaxes.online
Project-URL: Documentation, https://vectorharness.omniaxes.online/docs
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Intended Audience :: Developers
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: requests>=2.25.0
Requires-Dist: pydantic>=2.0.0
Provides-Extra: langchain
Requires-Dist: langchain>=0.1.0; extra == "langchain"

VectorHarness Python SDK 🧠

VectorHarness is a continuous, time-decay semantic memory engine for autonomous AI agents. Powered by OmniAxes Core.

Instead of writing custom Postgres vector extensions and managing background consolidation tasks, VectorHarness provides a high-velocity RAM queue and a fully transactional memory backend with one line of code.

Installation

pip install vectorharness


Quickstart

import os
from vectorharness import VectorHarnessClient

os.environ["VECTORHARNESS_API_KEY"] = "your_api_key_here"
client = VectorHarnessClient()

# 1. Ingest streaming memory (with JSON metadata)
client.add(
    session_id="chat-123", 
    actor="user", 
    raw_text="For the Apollo project, we are using PostgreSQL.",
    end_user_id="user_99",
    metadata={"project": "Apollo"}
)

# 2. Retrieve contextual memory (Hybrid Search + Time Decay)
results = client.search(
    query_text="What database are we using?",
    end_user_id="user_99",
    metadata_filter={"project": "Apollo"}
)

print(results["recommended_answer"])


LangChain Drop-In Replacement

If you are already using LangChain, you can give your agents infinite, auto-curating memory by swapping exactly one line of code.

pip install "vectorharness[langchain]"


from vectorharness import LangchainVectorHarnessMemory
from langchain.chains import ConversationChain
from langchain.llms import OpenAI

# Initialize VectorHarness Memory
vh_memory = LangchainVectorHarnessMemory(
    session_id="agent-session-42",
    end_user_id="user_99"
)

# Drop it into your existing LangChain agent!
conversation = ConversationChain(
    llm=OpenAI(temperature=0), 
    memory=vh_memory
)

conversation.predict(input="Hi, my name is Alex and I love Python.")
