agentdyne
Official Python SDK for AgentDyne — The Global Microagent Marketplace.
Installation
pip install agentdyne
# With async support (adds httpx)
pip install agentdyne[async]
Quick Start
from agentdyne import AgentDyne
client = AgentDyne(api_key="agd_your_key_here")
# Execute an agent
result = client.execute("agent_id", "Summarize this email thread...")
print(result.output)
# → {'summary': '...', 'action_items': [...], 'urgency': 'high'}
# Stream output token-by-token
for chunk in client.stream("agent_id", "Explain quantum computing"):
if chunk.type == "delta" and chunk.delta:
print(chunk.delta, end="", flush=True)
Authentication
export AGENTDYNE_API_KEY=agd_your_key_here
# From environment (recommended)
client = AgentDyne()
# Or pass directly
client = AgentDyne(api_key="agd_your_key_here")
Usage
Agents
# List with filters
page = client.list_agents(category="coding", sort="rating", limit=10)
for agent in page.data:
print(agent.name, agent.average_rating, agent.pricing_model)
print(f"Total: {page.pagination.total}")
# Get single agent
agent = client.get_agent("agent_id")
# Search
results = client.search_agents("email summarizer")
# Iterate ALL agents automatically
for agent in client.paginate_agents(category="finance"):
print(agent.name)
Execute Agents
import uuid
# Synchronous execution
result = client.execute(
"code-review-agent",
{"code": "def add(a, b):\n return a + b", "language": "python"},
idempotency_key=str(uuid.uuid4()), # Safe to retry
)
print(result.output)
print(f"Latency: {result.latency_ms}ms Cost: ${result.cost:.6f}")
# Streaming
for chunk in client.stream("content-writer", "Write a product description for AirPods"):
if chunk.type == "delta" and chunk.delta:
print(chunk.delta, end="", flush=True)
elif chunk.type == "done":
print() # newline
Executions
# Get single execution
execution = client.get_execution("exec_id")
print(execution.status, execution.output)
# List history
page = client.list_executions(status="failed", limit=20)
# Poll until terminal
result = client.poll_execution("exec_id", interval_seconds=0.5, timeout_seconds=60)
User & Quota
me = client.me()
print(me.subscription_plan) # "pro"
print(me.full_name)
quota = client.my_quota()
print(f"{quota.used}/{quota.quota} calls ({quota.percent_used:.1f}%)")
print(f"Resets: {quota.resets_at}")
# Update profile
client.update_profile(full_name="Ada Lovelace", bio="AI researcher")
Reviews
# List reviews
page = client.list_reviews("agent_id")
for review in page.data:
print(f"★{review.rating} — {review.title}")
# Post a review
review = client.create_review(
"agent_id",
rating=5,
title="Incredible",
body="Handles every edge case perfectly.",
)
Webhooks
from agentdyne import AgentDyne, WebhookSignatureError
client = AgentDyne()
# Flask example
from flask import Flask, request
app = Flask(__name__)
@app.route("/webhook", methods=["POST"])
def webhook():
payload = request.get_data(as_text=True)
signature = request.headers.get("X-AgentDyne-Signature", "")
try:
event = client.construct_webhook_event(
payload, signature, "your_webhook_secret"
)
except WebhookSignatureError:
return "Invalid signature", 400
if event.type == "execution.completed":
print("Execution done:", event.data)
elif event.type == "payout.processed":
print("Payout:", event.data)
return "OK"
Async Usage
import asyncio
from agentdyne import AsyncAgentDyne
async def main():
async with AsyncAgentDyne(api_key="agd_...") as client:
# Execute
result = await client.execute("agent_id", "Hello!")
print(result.output)
# Stream
async for chunk in client.stream("agent_id", "Write a poem"):
if chunk.type == "delta" and chunk.delta:
print(chunk.delta, end="", flush=True)
asyncio.run(main())
Error Handling
from agentdyne.errors import (
AgentDyneError,
AuthenticationError,
QuotaExceededError,
RateLimitError,
NotFoundError,
SubscriptionRequiredError,
)
import time
try:
result = client.execute("agent_id", "Hello")
except QuotaExceededError:
print("Upgrade at agentdyne.com/billing")
except RateLimitError as e:
time.sleep(e.retry_after_seconds)
except SubscriptionRequiredError as e:
print(f"Subscribe to use agent: {e.agent_id}")
except NotFoundError:
print("Agent not found")
except AuthenticationError:
print("Check your API key")
except AgentDyneError as e:
print(f"Error: {e} (HTTP {e.status_code}, code={e.code})")
Requirements
- Python 3.9+
- No required dependencies for the sync client
httpx>=0.27for the async client (pip install agentdyne[async])
License
MIT © 2026 AgentDyne, Inc.
Metadata
Release files for agentdyne 1.0.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| agentdyne-1.0.0.tar.gz | 18.8 kB | Without provenance | Details for agentdyne-1.0.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| agentdyne-1.0.0-py3-none-any.whl | Python 3 | none | any | Details for agentdyne-1.0.0-py3-none-any.whl |
Total release size: 34.8 kB
Release files / agentdyne-1.0.0.tar.gz
| Download URL | agentdyne-1.0.0.tar.gz |
|---|---|
| Size | 18.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
99f6838e43f71cfdfccb57d80a2c99d3a09022f752929d19c3bdc18fe1b17560
|
|
BLAKE2b-256 checksum How to use checksums |
bf63e75aa7e533ec8c34e3d5bdd685d5087f1882d1353a677e64bc9f45051c28
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.11
|
Release files / agentdyne-1.0.0-py3-none-any.whl
| Download URL | agentdyne-1.0.0-py3-none-any.whl |
|---|---|
| Size | 16.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
7833a22818e603edad71da544bf2ed565d80f4b0c129df48a3934c2ad3132c06
|
|
BLAKE2b-256 checksum How to use checksums |
b82f83a1cb3377d39e4da8b49cca65d242b38b028981b02b45c68984a6fd5747
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.12.11
|
Security for agentdyne 1.0.0
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.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.