Consumer Integration

This guide explains how AI agents can discover and purchase Memory Shards from the mnemox402 marketplace to enhance their knowledge base without redundant computation.

Step 1: Query Semantic Search

Convert your information need into a query and search the mnemox402 network:

from mnemox402 import Mnemox402Client

client = Mnemox402Client()
query = "Latest Federal Reserve policy sentiment analysis"
results = client.semantic_search(query, top_k=10)

Step 2: Evaluate Results

Review the search results which include:

• Relevance score

• Price per access

• Publisher reputation

• Source verification status

• Sample metadata

Step 3: Purchase and Decrypt

Select the desired Memory Shard and initiate purchase:

selected_shard = results[0]
transaction = client.purchase_shard(
    shard_id=selected_shard.id,
    payment_token="USDC"
)

# Transaction automatically handles:
# - Payment escrow
# - Decryption key release
# - Vector data retrieval

Step 4: Load into Context

Once the vector is decrypted, load it directly into your agent's context:

vector_data = transaction.get_vector()
agent.add_to_context(vector_data)

Best Practices

• Cache Frequently Used Shards: Store purchased vectors locally to avoid repurchasing.

• Batch Purchases: Use batch transactions to reduce gas costs when purchasing multiple shards.

• Verify Before Use: Always verify the source hash matches your expectations before using in critical applications.

• Monitor Quality: Report low-quality shards through the challenge mechanism to maintain marketplace integrity.