Python SDK for Vouchstone — AI engineers you can vouch for
Project description
Vouchstone SDK
Python SDK for building AI agents on the Vouchstone platform — the first accountable AI engineering firm.
Install
pip install vouchstone-sdk
# With vector memory (ChromaDB)
pip install vouchstone-sdk[vector]
# With graph memory (Neo4j)
pip install vouchstone-sdk[graph]
# Everything
pip install vouchstone-sdk[all]
Quick Start
from vouchstone_sdk import Agent, AgentConfig, Message
class MyAgent(Agent):
async def run(self, message: Message, context: dict):
relevant_docs = context.get("semantic", [])
history = context.get("episodic", [])
response = await self.llm.complete(
system=f"You are {self.config.name}. Use the context to help the user.",
messages=[{"role": "user", "content": message.content}],
context=relevant_docs,
)
return AgentResponse(content=response)
config = AgentConfig(
name="Data Migration Agent",
model="claude-sonnet-4-20250514",
semantic_memory=True,
episodic_memory=True,
procedural_memory=True,
)
agent = MyAgent(config)
await agent.initialize(vector_db_url="http://chromadb:8000")
response = await agent.process(Message(content="Migrate our PostgreSQL to Snowflake"))
5-Layer Memory Stack
Each agent has access to a biologically-inspired memory architecture:
| Layer | Name | Storage | Purpose |
|---|---|---|---|
| 1 | Working | Redis | Current turn context (resets per session) |
| 2 | Episodic | PostgreSQL | Cross-session traces (append-only) |
| 3 | Semantic | ChromaDB | Entity store (vector search) |
| 4 | Procedural | Neo4j | Learned skills (versioned DAG) |
| 5 | Meta | Control Plane | Decay, dedup, compress, forget |
Control Plane Client
from vouchstone_sdk import VouchstoneClient
client = VouchstoneClient(
api_url="https://api.vouchstone.ai",
api_key="your-api-key",
tenant_id="your-tenant-id",
)
# List agents
agents = await client.list_agents()
# Execute an agent
result = await client.execute(agent_id="...", input={"content": "Hello"})
License
Proprietary — Copyright (c) 2026 Vouchstone Inc. All Rights Reserved.
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
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file vouchstone_sdk-1.0.0.tar.gz.
File metadata
- Download URL: vouchstone_sdk-1.0.0.tar.gz
- Upload date:
- Size: 9.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.4
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
b474644c42c64e49cb6225ee37bf4e415f927e2da4cd641c4955d518f94154bc
|
|
| MD5 |
dc925132e97c19011c5bad70641409c0
|
|
| BLAKE2b-256 |
635a4c8dcce6fd19b753b5f2c393c58d4c692d9bfdf08cb777e6ff9381517938
|
File details
Details for the file vouchstone_sdk-1.0.0-py3-none-any.whl.
File metadata
- Download URL: vouchstone_sdk-1.0.0-py3-none-any.whl
- Upload date:
- Size: 9.3 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.4
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
a397b29ce01091f3155b527ffeb4a701dfdb8dcaf582d292a69bafac59a3c2c9
|
|
| MD5 |
bf7f6e4f360faabe6f33d3f338f0cda8
|
|
| BLAKE2b-256 |
10733c1f5efb481e63c11dd80d75f5d56f2327f5b67cf73c54bb04f5fa06c290
|