Skip to main content

piqrypt-autoGen-integration

Standard: AISS v2.0 · Full stack: piqrypt.com

Verifiable AI Agent Memory_Cryptographic audit trail for Microsoft AutoGen agents.

PyPI Downloads License: MIT PiQrypt

Every reply, code execution, and multi-agent conversation — signed, hash-chained, tamper-proof.

pip install piqrypt-autogen

Quickstart

from piqrypt_autogen import AuditedAssistant, AuditedUserProxy

# Drop-in replacements for AutoGen agents
assistant = AuditedAssistant(
    name="assistant",
    llm_config={"model": "gpt-4o"},
    identity_file="assistant.json"    # PiQrypt identity
)

user_proxy = AuditedUserProxy(
    name="user_proxy",
    human_input_mode="NEVER",
    identity_file="proxy.json"
)

# Normal AutoGen conversation — every reply is stamped
user_proxy.initiate_chat(
    assistant,
    message="Analyze the Q4 financial report and flag anomalies."
)

# Export tamper-proof audit trail
assistant.export_audit("q4-analysis-audit.json")
# $ piqrypt verify q4-analysis-audit.json

Multi-agent group chat

from autogen import GroupChat
from piqrypt_autogen import AuditedAssistant, AuditedUserProxy, AuditedGroupChat

researcher = AuditedAssistant(
    name="researcher",
    llm_config=llm_config,
    identity_file="researcher.json"
)

analyst = AuditedAssistant(
    name="analyst",
    llm_config=llm_config,
    identity_file="analyst.json"
)

user_proxy = AuditedUserProxy(
    name="user",
    human_input_mode="NEVER"
)

group_chat = GroupChat(
    agents=[user_proxy, researcher, analyst],
    messages=[],
    max_round=10
)

manager = AuditedGroupChat(
    groupchat=group_chat,
    llm_config=llm_config,
    identity_file="manager.json"
)

user_proxy.initiate_chat(manager, message="Research AI Act compliance requirements")

# Every agent stamped independently — full end-to-end audit
manager.export_audit("group-chat-audit.json")

Stamp a third-party agent

from piqrypt_autogen import stamp_reply

# Can't subclass? Wrap the method directly
agent = SomeThirdPartyAgent(...)
agent.generate_reply = stamp_reply(
    identity_file="my-agent.json"
)(agent.generate_reply)

Stamp an entire conversation post-hoc

from piqrypt_autogen import stamp_conversation

# After a conversation completes
conv_hash = stamp_conversation(
    messages=user_proxy.chat_messages[assistant],
    private_key=private_key,
    agent_id=agent_id,
    conversation_id="q4-analysis-2026-02"
)

What gets stamped

Event Agent When
agent_initialized AssistantAgent On creation
proxy_initialized UserProxy On creation
groupchat_initialized GroupChatManager On creation
reply_generated AssistantAgent After each reply
proxy_reply UserProxy After each proxy reply
code_executed UserProxy After code block execution
groupchat_start GroupChatManager Before chat runs
groupchat_complete GroupChatManager After chat finishes

All events Ed25519-signed, SHA-256 hash-chained. Raw messages never stored — only hashes.


Verify

piqrypt verify autogen-audit.json
# ✅ Chain integrity verified — 18 events, 0 forks

piqrypt search --type code_executed
# All code executions with timestamps


Metadata

Release files for piqrypt-autoGen-integration 1.1.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for piqrypt-autoGen-integration 1.1.0
File Size Uploaded
piqrypt_autogen_integration-1.1.0.tar.gz 3.7 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for piqrypt-autoGen-integration 1.1.0
File Interpreter ABI Platform
piqrypt_autogen_integration-1.1.0-py3-none-any.whl Python 3 none any Details

Total release size: 7.6 kB

Release files / piqrypt_autogen_integration-1.1.0.tar.gz

Download URL piqrypt_autogen_integration-1.1.0.tar.gz
Size 3.7 kB
Tags Source
SHA-256 checksum
How to use checksums
1e215385990cdf1afd77ceaea8b6755f969389e78c81747f1418c466dca6b4c4
BLAKE2b-256 checksum
How to use checksums
8bdb0bd19cd230060aef7ad383204849a435865c6d41700ac81eb86db1271f05
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Apr 19, 2026.

Transparency log

Release files / piqrypt_autogen_integration-1.1.0-py3-none-any.whl

Download URL piqrypt_autogen_integration-1.1.0-py3-none-any.whl
Size 3.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
6a97491be7fe34f541eb222ceddfe1d60c28b37ab50ab51bef36a41584bcbba5
BLAKE2b-256 checksum
How to use checksums
73e784c971e803438fe082c7e7b96118205fa674aa1d668770bf226bb9c39118
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/6.1.0 CPython/3.13.12

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Apr 19, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

1.1.0 This release

2 release files

1.0.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page