nahook

Official Python SDK for the Nahook webhook platform.

Two classes, one package:

Class	Purpose	Auth
NahookClient	Send and trigger webhook events	API key (nhk_us_...)
NahookManagement	Manage endpoints, event types, apps	Management token (nhm_...)

Requirements

• Python 3.9+
• httpx (installed automatically)

Installation

pip install nahook

---

NahookClient

Send webhooks to specific endpoints or fan-out by event type.

Setup

from nahook import NahookClient

# Simple
client = NahookClient("nhk_us_...")

# With options
client = NahookClient("nhk_us_...", retries=3, timeout=5_000)
# retries: default 0 (no retries)
# timeout: default 30_000ms

Configuration

The SDK automatically routes requests to the correct regional API based on your API key prefix (nhk_us_... -> US, nhk_eu_... -> EU, nhk_ap_... -> Asia Pacific). No configuration needed.

To override the base URL (for testing or local development):

client = NahookClient("nhk_us_...", base_url="http://localhost:3001")

For unit tests, mock the SDK client at the dependency injection boundary. For integration tests, override the base URL to point at a local server.

Graceful shutdown — with statement and close()

NahookClient and NahookManagement are context managers — the idiomatic Python pattern is with, which closes the underlying httpx.Client's connection pool on exit:

with NahookClient("nhk_us_...") as client:
    client.send("ep_abc123", payload={"orderId": "123", "status": "paid"})
# httpx.Client.close() automatically called here

If with doesn't fit your lifecycle (e.g., a long-lived module-scope client), call close() explicitly during graceful shutdown:

import atexit

client = NahookClient("nhk_us_...")
atexit.register(client.close)

# ... use client across the process lifetime ...

close() drains pooled connections and is safe to call multiple times. The same pattern works on NahookManagement. Skipping close() is fine in short-lived scripts — the OS reaps sockets on process exit — but matters for test harnesses, graceful shutdown handlers, or any process that recycles clients during its lifetime.

Send to a specific endpoint

result = client.send("ep_abc123", {
    "orderId": "123",
    "status": "paid",
}, idempotency_key="order-123-paid")  # optional, auto-generated UUID if omitted

# {"deliveryId": "del_...", "idempotencyKey": "order-123-paid", "status": "accepted"}

Fan-out by event type

result = client.trigger("order.paid", {
    "orderId": "123",
    "status": "paid",
}, metadata={"region": "us-east-1"})  # optional

# {"eventTypeId": "evt_...", "deliveryIds": ["del_..."], "status": "accepted"}

Batch operations

# Send to multiple endpoints (max 20 items)
batch = client.send_batch([
    {"endpointId": "ep_abc", "payload": {"orderId": "123"}},
    {"endpointId": "ep_def", "payload": {"orderId": "456"}},
])

# Fan-out multiple event types (max 20 items)
fan_out = client.trigger_batch([
    {"eventType": "order.paid", "payload": {"orderId": "123"}},
    {"eventType": "order.shipped", "payload": {"orderId": "456"}},
])

# Results: 202 (all succeed) or 207 (mixed)
for item in batch["items"]:
    if "error" in item:
        print(f"Item {item['index']} failed: {item['error']['code']}")

Retry behavior

Retries are opt-in via the retries constructor parameter. When enabled:

• Strategy: Exponential backoff with full jitter
• Delays: 500ms base, 10s max
• Retryable: 5xx, 429 (respects Retry-After), network errors, timeouts
• Non-retryable: 400, 401, 403, 404, 409, 413
• Safe by design: Idempotency keys are always sent, making retries safe

---

NahookManagement

Programmatically manage your Nahook workspace resources.

Setup

from nahook import NahookManagement

# Simple
mgmt = NahookManagement("nhm_...")

# With options
mgmt = NahookManagement("nhm_...", timeout=10_000)
# Note: retries are not supported for management calls

Endpoints

result = mgmt.endpoints.list("ws_abc")
endpoints = result["data"]

endpoint = mgmt.endpoints.create("ws_abc",
    url="https://example.com/webhooks",
    description="Production webhook",
    type_="webhook",  # "webhook" | "slack"
)

endpoint = mgmt.endpoints.get("ws_abc", "ep_123")

mgmt.endpoints.update("ws_abc", "ep_123",
    description="Updated",
    is_active=False,
)

mgmt.endpoints.delete("ws_abc", "ep_123")

Event Types

result = mgmt.event_types.list("ws_abc")

event_type = mgmt.event_types.create("ws_abc",
    name="order.paid",
    description="Fired when an order is paid",
)

event_type = mgmt.event_types.get("ws_abc", "evt_123")

mgmt.event_types.update("ws_abc", "evt_123",
    description="Updated description",
)

mgmt.event_types.delete("ws_abc", "evt_123")

Applications

result = mgmt.applications.list("ws_abc", limit=50, offset=0)

app = mgmt.applications.create("ws_abc",
    name="Acme Corp",
    external_id="acme-123",
    metadata={"tier": "pro"},
)

app = mgmt.applications.get("ws_abc", "app_123")

mgmt.applications.update("ws_abc", "app_123", name="Acme Inc")

mgmt.applications.delete("ws_abc", "app_123")

# Endpoints scoped to an application
result = mgmt.applications.list_endpoints("ws_abc", "app_123")
ep = mgmt.applications.create_endpoint("ws_abc", "app_123",
    url="https://acme.com/webhooks",
)

Subscriptions

result = mgmt.subscriptions.list("ws_abc", "ep_123")

mgmt.subscriptions.create("ws_abc", "ep_123", event_type_ids=["evt_456"])

mgmt.subscriptions.delete("ws_abc", "ep_123", "evt_456")

Environments

result = mgmt.environments.list("ws_abc")

env = mgmt.environments.create("ws_abc",
    name="Staging",
    slug="staging",
)

env = mgmt.environments.get("ws_abc", "env_123")

mgmt.environments.update("ws_abc", "env_123", name="Pre-production")

mgmt.environments.delete("ws_abc", "env_123")

Event Type Visibility

Control which event types are visible per environment.

result = mgmt.environments.list_event_type_visibility("ws_abc", "env_123")

vis = mgmt.environments.set_event_type_visibility("ws_abc", "env_123", "evt_456",
    published=True,
)
# {"eventTypeId": "evt_456", "eventTypeName": "order.paid", "published": True}

Deliveries

Read access to a workspace's webhook deliveries — list, fetch, and inspect attempts. There is no create/update/delete on this resource.

# Page through an endpoint's deliveries (newest-first).
result = mgmt.deliveries.list("ws_abc", "ep_123", limit=50)
for delivery in result["data"]:
    print(delivery["id"], delivery["status"])

# next_cursor is opaque — pass it through verbatim to fetch the next page.
# It is None when there are no more pages.
if result["next_cursor"] is not None:
    next_page = mgmt.deliveries.list(
        "ws_abc", "ep_123", limit=50, cursor=result["next_cursor"]
    )

# Filter by status.
failed = mgmt.deliveries.list("ws_abc", "ep_123", status="failed")

# Fetch a single delivery's metadata.
delivery = mgmt.deliveries.get("ws_abc", "del_abc")
print(delivery["status"], delivery["totalAttempts"])

# Fetch with the stored payload envelope. The envelope's ``status`` carries
# the access-level reality — only ``"available"`` payloads include ``data``.
delivery = mgmt.deliveries.get("ws_abc", "del_abc", include_payload=True)
envelope = delivery["payload"]
if envelope["status"] == "available":
    print(envelope["data"], envelope["contentType"])
elif envelope["status"] == "forbidden":
    print("Workspace plan does not include payload storage")
elif envelope["status"] == "processing":
    print("Delivery still in flight — try again shortly")
elif envelope["status"] == "not_found":
    print("No stored payload for this delivery")
elif envelope["status"] == "error":
    print("Transient storage failure")

# List a delivery's attempts (chronological, oldest first).
attempts = mgmt.deliveries.get_attempts("ws_abc", "del_abc")
for attempt in attempts:
    print(
        attempt["attemptNumber"],
        attempt["status"],
        attempt["responseStatusCode"],
    )

Portal Sessions

session = mgmt.portal_sessions.create("ws_abc", "app_123",
    metadata={"userId": "user-456"},
)
# session["url"]        -> redirect end-user here
# session["code"]       -> one-time exchange code
# session["expiresAt"]  -> expiration timestamp

---

Error Handling

All SDK errors extend NahookError. Three specific types cover every failure mode:

from nahook import NahookAPIError, NahookNetworkError, NahookTimeoutError

try:
    client.send("ep_abc", {"key": "value"})
except NahookAPIError as err:
    # API returned an error response
    print(err.status)        # 404
    print(err.code)          # "not_found"
    print(str(err))          # "Endpoint not found"
    print(err.retry_after)   # seconds (on 429s)

    # Convenience checks
    err.is_retryable       # True for 5xx, 429
    err.is_auth_error      # True for 401, 403 (token_disabled)
    err.is_not_found       # True for 404
    err.is_rate_limited    # True for 429
    err.is_validation_error  # True for 400
except NahookNetworkError as err:
    print(err.cause)  # original httpx error
except NahookTimeoutError as err:
    print(err.timeout_ms)  # timeout that was exceeded

---

Webhook Verification

Nahook signs outgoing deliveries using the Standard Webhooks specification. Use the standardwebhooks package to verify incoming webhooks:

pip install standardwebhooks

from standardwebhooks import Webhook

wh = Webhook("whsec_MfKQ9r8GKYqr...")

try:
    payload = wh.verify(request.body, request.headers)
    # Verified and safe to use
except Exception:
    # Invalid signature
    pass

The signing secret (whsec_...) is available in your Nahook Dashboard endpoint settings.

---

Development

pip install -e ".[dev]"   # install with dev dependencies
pytest                     # run tests

License

MIT