Knowledge Management

Sypho provides structured data storage for agents.

Storage Type

Structured Data

JSON document storage with:

• Namespaces - Logical grouping
• Keys - Unique identifiers
• Data - JSON documents
• Metadata - Queryable attributes
• Versioning - Full history tracking

Knowledge Scope

Every run has a knowledge scope that defines data access:

{
  "structured_data_namespaces": {
    "app-data": {
      "keys": ["*"],               // All keys
      "permission": "read_write"   // Full access
    },
    "shared-config": {
      "keys": ["public:*"],        // Only public:* keys
      "permission": "read"          // Read-only
    }
  }
}

Permissions

• read - Get, query, list operations
• read_write - All operations (save, delete)

Platform Tools

Access via context.call_tool():

Tool	Description
structured_data_save	Create/update entry
structured_data_get	Fetch by key
structured_data_query	Search by metadata
structured_data_delete	Remove entry
structured_data_list_keys	List keys in namespace

Usage Example

from sypho_sdk import entrypoint, AgentContext

@entrypoint
async def main(input: dict, context: AgentContext):
    # Save data
    await context.call_tool("structured_data_save", {
        "namespace": "users",
        "key": "user:123",
        "data": {"name": "Alice", "age": 30},
        "metadata": {"type": "user", "region": "us"}
    })

    # Query by metadata
    results = await context.call_tool("structured_data_query", {
        "namespace": "users",
        "filters": {
            "metadata.type": "user",
            "metadata.region": "us"
        }
    })

    return results

Namespaces

Namespaces provide logical isolation:

• Per-agent: my-agent-data
• Shared: shared-config
• User-specific: user-{user_id}-data

Next Steps

• Structured Data
• Data Agent Example