Skip to content

Commit 65ce01b

Browse files
constantiniusconstantinius
authored
feat(python): add docs for Pydantic AI integration (#15177)
1 parent 015712e commit 65ce01b

File tree

3 files changed

+261
-1
lines changed

3 files changed

+261
-1
lines changed

‎docs/platforms/python/integrations/index.mdx‎

Lines changed: 3 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -48,7 +48,9 @@ The Sentry SDK uses integrations to hook into the functionality of popular libra
4848
| <LinkWithPlatformIcon platform="langchain" label="LangChain" url="/platforms/python/integrations/langchain" /> | ✓ |
4949
| <LinkWithPlatformIcon platform="langgraph" label="LangGraph" url="/platforms/python/integrations/langgraph" /> | ✓ |
5050
| <LinkWithPlatformIcon platform="litellm" label="LiteLLM" url="/platforms/python/integrations/litellm" /> | |
51-
| <LinkWithPlatformIcon platform="mcp" label="MCP (Model Context Protocol)" url="/platforms/python/integrations/mcp" /> | |
51+
| <LinkWithPlatformIcon platform="pydantic-ai" label="Pydantic AI" url="/platforms/python/integrations/pydantic-ai" /> | |
52+
| <LinkWithPlatformIcon platform="mcp" label="MCP (Model Context Protocol)" url="/platforms/python/integrations/mcp" /> | |
53+
5254

5355
### Data Processing
5456

‎docs/platforms/python/integrations/pydantic-ai/index.mdx‎

Lines changed: 257 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,257 @@
1+
---
2+
title: Pydantic AI
3+
description: "Learn about using Sentry for Pydantic AI."
4+
---
5+
6+
<Alert title="Beta">
7+
8+
The support for **Pydantic AI** is in beta. Please test locally before using in production.
9+
10+
</Alert>
11+
12+
This integration connects Sentry with the [Pydantic AI](https://ai.pydantic.dev/) library.
13+
The integration has been confirmed to work with Pydantic AI version 1.0.0+.
14+
15+
Once you've installed this integration, you can use [Sentry AI Agents Insights](https://sentry.io/orgredirect/organizations/:orgslug/insights/ai/agents/), a Sentry dashboard that helps you understand what's going on with your AI agents.
16+
17+
Sentry AI Agents monitoring will automatically collect information about agents, tools, prompts, tokens, and models.
18+
19+
## Install
20+
21+
Install `sentry-sdk` from PyPI:
22+
23+
```bash {tabTitle:pip}
24+
pip install "sentry-sdk"
25+
```
26+
27+
```bash {tabTitle:uv}
28+
uv add "sentry-sdk"
29+
```
30+
31+
## Configure
32+
33+
Add `PydanticAIIntegration()` to your `integrations` list:
34+
35+
```python {tabTitle:OpenAI}
36+
import sentry_sdk
37+
from sentry_sdk.integrations.pydantic_ai import PydanticAIIntegration
38+
from sentry_sdk.integrations.openai import OpenAIIntegration
39+
40+
sentry_sdk.init(
41+
dsn="___PUBLIC_DSN___",
42+
traces_sample_rate=1.0,
43+
# Add data like LLM and tool inputs/outputs;
44+
# see https://docs.sentry.io/platforms/python/data-management/data-collected/ for more info
45+
send_default_pii=True,
46+
integrations=[
47+
PydanticAIIntegration(),
48+
],
49+
# Disable the OpenAI integration to avoid double reporting of chat spans
50+
disabled_integrations=[OpenAIIntegration()],
51+
)
52+
```
53+
54+
```python {tabTitle:Anthropic}
55+
import sentry_sdk
56+
from sentry_sdk.integrations.pydantic_ai import PydanticAIIntegration
57+
58+
sentry_sdk.init(
59+
dsn="___PUBLIC_DSN___",
60+
traces_sample_rate=1.0,
61+
# Add data like LLM and tool inputs/outputs;
62+
# see https://docs.sentry.io/platforms/python/data-management/data-collected/ for more info
63+
send_default_pii=True,
64+
integrations=[
65+
PydanticAIIntegration(),
66+
],
67+
)
68+
```
69+
70+
<Alert level="warning">
71+
72+
When using Pydantic AI with OpenAI models, you must disable the OpenAI integration to avoid double reporting of chat spans. Add `disabled_integrations=[OpenAIIntegration()]` to your `sentry_sdk.init()` call as shown in the OpenAI tab above.
73+
74+
</Alert>
75+
76+
## Verify
77+
78+
Verify that the integration works by running an AI agent. The resulting data should show up in your AI Agents Insights dashboard. In this example, we're creating a customer support agent that analyzes customer inquiries and can optionally look up order information using a tool.
79+
80+
```python {tabTitle:OpenAI}
81+
import asyncio
82+
83+
import sentry_sdk
84+
from sentry_sdk.integrations.pydantic_ai import PydanticAIIntegration
85+
from sentry_sdk.integrations.openai import OpenAIIntegration
86+
from pydantic_ai import Agent, RunContext
87+
from pydantic import BaseModel
88+
89+
class SupportResponse(BaseModel):
90+
message: str
91+
sentiment: str
92+
requires_escalation: bool
93+
94+
support_agent = Agent(
95+
'openai:gpt-4o-mini',
96+
name="Customer Support Agent",
97+
system_prompt=(
98+
"You are a helpful customer support agent. Analyze customer inquiries, "
99+
"provide helpful responses, and determine if escalation is needed. "
100+
"If the customer mentions an order number, use the lookup tool to get details."
101+
),
102+
result_type=SupportResponse,
103+
)
104+
105+
@support_agent.tool
106+
async def lookup_order(ctx: RunContext[None], order_id: str) -> dict:
107+
"""Look up order details by order ID.
108+
109+
Args:
110+
ctx: The context object.
111+
order_id: The order identifier.
112+
113+
Returns:
114+
Order details including status and tracking.
115+
"""
116+
# In a real application, this would query a database
117+
return {
118+
"order_id": order_id,
119+
"status": "shipped",
120+
"tracking_number": "1Z999AA10123456784",
121+
"estimated_delivery": "2024-03-15"
122+
}
123+
124+
async def main() -> None:
125+
sentry_sdk.init(
126+
dsn="___PUBLIC_DSN___",
127+
traces_sample_rate=1.0,
128+
# Add data like LLM and tool inputs/outputs;
129+
# see https://docs.sentry.io/platforms/python/data-management/data-collected/ for more info
130+
send_default_pii=True,
131+
integrations=[
132+
PydanticAIIntegration(),
133+
],
134+
# Disable the OpenAI integration to avoid double reporting of chat spans
135+
disabled_integrations=[OpenAIIntegration()],
136+
)
137+
138+
result = await support_agent.run(
139+
"Hi, I'm wondering about my order #ORD-12345. When will it arrive?"
140+
)
141+
print(result.data)
142+
143+
if __name__ == "__main__":
144+
asyncio.run(main())
145+
```
146+
147+
```python {tabTitle:Anthropic}
148+
import asyncio
149+
150+
import sentry_sdk
151+
from sentry_sdk.integrations.pydantic_ai import PydanticAIIntegration
152+
from pydantic_ai import Agent, RunContext
153+
from pydantic import BaseModel
154+
155+
class SupportResponse(BaseModel):
156+
message: str
157+
sentiment: str
158+
requires_escalation: bool
159+
160+
support_agent = Agent(
161+
'anthropic:claude-3-5-sonnet-latest',
162+
name="Customer Support Agent",
163+
system_prompt=(
164+
"You are a helpful customer support agent. Analyze customer inquiries, "
165+
"provide helpful responses, and determine if escalation is needed. "
166+
"If the customer mentions an order number, use the lookup tool to get details."
167+
),
168+
result_type=SupportResponse,
169+
)
170+
171+
@support_agent.tool
172+
async def lookup_order(ctx: RunContext[None], order_id: str) -> dict:
173+
"""Look up order details by order ID.
174+
175+
Args:
176+
ctx: The context object.
177+
order_id: The order identifier.
178+
179+
Returns:
180+
Order details including status and tracking.
181+
"""
182+
# In a real application, this would query a database
183+
return {
184+
"order_id": order_id,
185+
"status": "shipped",
186+
"tracking_number": "1Z999AA10123456784",
187+
"estimated_delivery": "2024-03-15"
188+
}
189+
190+
async def main() -> None:
191+
sentry_sdk.init(
192+
dsn="___PUBLIC_DSN___",
193+
traces_sample_rate=1.0,
194+
# Add data like LLM and tool inputs/outputs;
195+
# see https://docs.sentry.io/platforms/python/data-management/data-collected/ for more info
196+
send_default_pii=True,
197+
integrations=[
198+
PydanticAIIntegration(),
199+
],
200+
)
201+
202+
result = await support_agent.run(
203+
"Hi, I'm wondering about my order #ORD-12345. When will it arrive?"
204+
)
205+
print(result.data)
206+
207+
if __name__ == "__main__":
208+
asyncio.run(main())
209+
```
210+
211+
It may take a couple of moments for the data to appear in [sentry.io](https://sentry.io).
212+
213+
## Behavior
214+
215+
Data on the following will be collected:
216+
217+
- AI agents invocations
218+
- execution of tools
219+
- number of input and output tokens used
220+
- LLM models usage
221+
- model settings (temperature, max_tokens, etc.)
222+
223+
Sentry considers LLM and tool inputs/outputs as PII and doesn't include PII data by default. If you want to include the data, set `send_default_pii=True` in the `sentry_sdk.init()` call. To explicitly exclude prompts and outputs despite `send_default_pii=True`, configure the integration with `include_prompts=False` as shown in the Options section below.
224+
225+
## Options
226+
227+
By adding `PydanticAIIntegration` to your `sentry_sdk.init()` call explicitly, you can set options for `PydanticAIIntegration` to change its behavior:
228+
229+
```python
230+
import sentry_sdk
231+
from sentry_sdk.integrations.pydantic_ai import PydanticAIIntegration
232+
233+
sentry_sdk.init(
234+
# ...
235+
# Add data like inputs and responses;
236+
# see https://docs.sentry.io/platforms/python/data-management/data-collected/ for more info
237+
send_default_pii=True,
238+
integrations=[
239+
PydanticAIIntegration(
240+
include_prompts=False, # LLM and tool inputs/outputs will be not sent to Sentry, despite send_default_pii=True
241+
),
242+
],
243+
)
244+
```
245+
246+
You can pass the following keyword arguments to `PydanticAIIntegration()`:
247+
248+
- `include_prompts`:
249+
250+
Whether LLM and tool inputs and outputs should be sent to Sentry. Sentry considers this data personal identifiable data (PII) by default. If you want to include the data, set `send_default_pii=True` in the `sentry_sdk.init()` call. To explicitly exclude prompts and outputs despite `send_default_pii=True`, configure the integration with `include_prompts=False`.
251+
252+
The default is `True`.
253+
254+
## Supported Versions
255+
256+
- Pydantic AI: 1.0.0+
257+
- Python: 3.9+

‎docs/platforms/python/tracing/instrumentation/custom-instrumentation/ai-agents-module.mdx‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -19,6 +19,7 @@ The Python SDK supports automatic instrumentation for some AI libraries. We reco
1919
- <PlatformLink to="/integrations/langchain/">LangChain</PlatformLink>
2020
- <PlatformLink to="/integrations/langgraph/">LangGraph</PlatformLink>
2121
- <PlatformLink to="/integrations/litellm/">LiteLLM</PlatformLink>
22+
- <PlatformLink to="/integrations/pydantic-ai/">Pydantic AI</PlatformLink>
2223

2324
## Manual Instrumentation
2425

0 commit comments

Comments
 (0)