prompt = client.prompts.update(slug, options)
# Async
prompt = await client.prompts.aupdate(slug, options)
Parameters
| Parameter | Type | Required | Description |
|---|
slug | str | Yes | The prompt slug to update |
options | UpdatePromptOptions | Yes | Update options object |
options.name | str | No | New display name |
options.description | str | No | New description |
options.messages | list[PromptMessage] | No | New messages (creates new version) |
options.model | str | No | New default model |
options.provider | LLMProvider | No | New provider |
options.temperature | float | No | New default temperature |
options.max_output_tokens | int | No | New max output tokens |
options.top_p | float | No | New top-p value |
options.stop_sequences | list[str] | No | New stop sequences |
Response
Returns the updated Prompt object with the new version number if messages changed.
Examples
from tracia import UpdatePromptOptions
prompt = client.prompts.update("welcome-email", UpdatePromptOptions(
name="Welcome Email v2",
description="Updated welcome email template",
))
# Version stays the same when only metadata changes
print(prompt.current_version) # 1
Update Messages (Creates New Version)
from tracia import UpdatePromptOptions, PromptMessage
prompt = client.prompts.update("welcome-email", UpdatePromptOptions(
messages=[
PromptMessage(role="system", content="You write friendly, casual emails."),
PromptMessage(role="user", content="Write a welcome email for {{name}} at {{company}}."),
],
))
# Version incremented because messages changed
print(prompt.current_version) # 2
Update Model and Provider
prompt = client.prompts.update("welcome-email", UpdatePromptOptions(
model="claude-sonnet-4-20250514",
provider="anthropic",
temperature=0.8,
))
Versioning Behavior
Message changes automatically create a new version. This allows you to track changes over time and see which version was used for each span.
from tracia import CreatePromptOptions, UpdatePromptOptions, PromptMessage
# Version 1: Original messages
v1 = client.prompts.create(CreatePromptOptions(
slug="my-prompt",
name="My Prompt",
model="gpt-4o",
provider="openai",
messages=[PromptMessage(role="user", content="Hello {{name}}")],
))
print(v1.current_version) # 1
# Version 2: Updated messages
v2 = client.prompts.update("my-prompt", UpdatePromptOptions(
messages=[PromptMessage(role="user", content="Hi {{name}}!")],
))
print(v2.current_version) # 2
# Metadata update: No new version
v2b = client.prompts.update("my-prompt", UpdatePromptOptions(
name="My Updated Prompt",
))
print(v2b.current_version) # Still 2
Error Handling
from tracia import TraciaError, TraciaErrorCode
try:
prompt = client.prompts.update("welcome-email", UpdatePromptOptions(
name="New Name",
))
except TraciaError as error:
if error.code == TraciaErrorCode.NOT_FOUND:
print("Prompt does not exist")
elif error.code == TraciaErrorCode.INVALID_REQUEST:
print("Invalid request")
