attestify-sdk 0.1.0

Official Python SDK for Attestify OS — governed agent execution in a few lines.

attestify-sdk — Python

Add governed agent execution to any Python codebase in a few lines.

Install

pip install attestify-sdk

Or directly from the repo while the package is in early access:

pip install "git+https://github.com/attestifyagent/attestify-os.git#subdirectory=sdk/python"

Quick start

import os
from attestify import create_client

client = create_client(api_key=os.environ["ATTESTIFY_API_KEY"])

result = client.run_agent(
    input="Summarise the latest advances in quantum error correction.",
    agent_id="researcher-v2",
    session_id="my-session-001",
    options={"verify": True, "write_memory": True},
)

print(result.output)             # agent response text
print(result.receipt.run_id)     # unique run identifier
print(result.receipt.evidence)   # full EvidenceBundle dataclass
print(result.receipt.pricing)    # cost breakdown

Features

• create_client(api_key) — configure once, use anywhere
• client.run_agent(input, ...) — single call handles sessions, retries, idempotency
• Typed receipts — RunReceipt and EvidenceBundle dataclasses with full field coverage
• Auto-routing — omit agent_id and Attestify picks the best agent for the task
• Idempotency — pass idempotency_key to guarantee exactly-once execution
• Budget controls — pass constraints={"max_cost_usdc": 0.05} to cap spend per run
• Webhook support — pass options={"webhook_url": "..."} for async delivery
• Zero dependencies — stdlib only (urllib, json, uuid)

API Reference

create_client(api_key, base_url?, max_retries?, timeout_s?)

Parameter	Type	Default	Description
api_key	str	required	Your Attestify API key (ATTESTIFY_API_KEY)
base_url	str	https://attestify-os.vercel.app	Override the endpoint
max_retries	int	2	Retries on transient 5xx errors
timeout_s	float	60.0	Per-attempt timeout in seconds

---

client.run_agent(input, ...)

Parameter	Type	Description
input	str	Required. The task or intent
agent_id	str	Optional — omit to auto-route
session_id	str	Memory / conversation continuity ID
idempotency_key	str	Prevents duplicate charges for the same key
context	dict	Freeform context forwarded to the agent
constraints	dict	Budget / SLA constraints (see below)
options	dict	Runtime flags (see below)

Returns RunAgentResult(output, receipt, raw).

constraints fields:

constraints={
    "max_cost_usdc": 0.05,    # abort if estimated cost exceeds this
    "max_latency_ms": 10000,  # abort if estimated latency exceeds this
    "budget_id": "proj-abc",  # link to a named budget envelope
}

options fields:

options={
    "verify": True,                    # run output verification (default False)
    "write_memory": True,              # persist session memory (default False)
    "include_memory": True,            # inject prior memory into context
    "webhook_url": "https://...",      # async result delivery
}

---

client.get_receipt(loop_id)

Fetch a stored receipt by its loop_id.

receipt = client.get_receipt("loop_abc123")
print(receipt.verification)   # grade, score, output_hash
print(receipt.settlement)     # on-chain tx hash if x402 used

Returns RunReceipt.

---

RunReceipt fields

Field	Type	Description
run_id	str	Unique run identifier
loop_id	str	Persistent loop/session receipt ID
agent_id	str	Agent that handled the run
output	str	Agent response text
paid	bool	Whether x402 payment was used
subscription_used	bool	Whether an API key subscription covered this run
evidence	EvidenceBundle	Full governance evidence bundle
verification	dict | None	Verification grade, score, output hash
pricing	dict	Cost breakdown in USDC
settlement	dict | None	On-chain settlement details
receipt_url	str	Shareable receipt permalink

---

Available agents

agent_id	Description
researcher-v2	Deep research and synthesis
analyst-v1	Data analysis and structured output
coder-v1	Code generation and review
writer-v1	Long-form and structured writing
strategist-v1	Strategic planning and frameworks
support-v1	Customer support and triage
comedian-v1	Creative and entertainment tasks

Omit agent_id entirely to let Attestify auto-route based on your input.

Environment variables

ATTESTIFY_API_KEY=atst_live_...

Async usage

The client is synchronous by design (zero dependencies). For async workflows:

import asyncio
from attestify import create_client

client = create_client(api_key=os.environ["ATTESTIFY_API_KEY"])

async def main():
    result = await asyncio.to_thread(
        client.run_agent,
        input="Analyse Q2 revenue trends",
        agent_id="analyst-v1",
    )
    print(result.output)

asyncio.run(main())