Installation
Install with your package manager of choice:Quick-start: the Interaction API
The new interaction workflow is a three-step pattern:begin()
- creates an interaction object and logs the initial user input.- Update - optionally call
setProperty
,setProperties
, oraddAttachments
. finish()
- records the AI’s final output and closes the interaction.
Using Vercel AI SDK? If you’re using the Vercel AI SDK, you can use our easy integration here to automatically track AI events and traces.
It is currently in beta and we’d love your feedback while we continue to improve the experience!
Example: chat completion with the ai
SDK
Updating an interaction
You can update an interaction at any time usingsetProperty
, setProperties
, or addAttachments
.
Resuming an interaction
If you don’t have access to the interaction object that was returned frombegin()
, you can resume an interaction by calling resumeInteraction()
.
Interactions are subject to the global 1 MB event limit; oversized payloads will be truncated. Contact us if you have custom requirements.
Single-shot tracking (legacy trackAi
)
If your interaction is atomic (e.g. “user asked, model answered” in one function) you can still call trackAi()
directly:
Heads‑up: We recommend migrating tobegin()
→finish()
for all new code so you gain partial‑event buffering, tracing helpers, and upcoming features such as automatic token counts.
Tracking Signals (feedback)
Signals capture explicit or implicit quality ratings on an earlier AI event. UsetrackSignal()
with the same eventId
you used in begin()
or trackAi()
.
Parameter | Type | Description |
---|---|---|
eventId | string | The ID of the AI event you’re evaluating |
name | "thumbs_up", "thumbs_down" , string | Name of the signal (e.g. "thumbs_up" ) |
type | "default", "feedback", "edit" | Optional, defaults to "default" |
comment | string | For feedback signals |
after | string | For edit signals – the user’s final content |
sentiment | "POSITIVE", "NEGATIVE" | indicates whether the signal is positive (default is NEGATIVE) |
…others | See API reference |
Attachments
Attachments allow you to include context from the user or that hte model outputted. These could be documents, generated images, code, or even an entire web page. They work the same way inbegin()
interactions and in single‑shot trackAi
calls.
Each attachment is an object with the following properties:
type
(string): The type of attachment. Can be “code”, “text”, “image”, or “iframe”.name
(optional string): A name for the attachment.value
(string): The content or URL of the attachment.role
(string): Either “input” or “output”, indicating whether the attachment is part of the user input or AI output.language
(optional string): For code attachments, specifies the programming language.
code
, text
, image
, iframe
.
Identifying users
PII redaction
Read more on how Raindrop handles privacy and PII redaction here. Note that this doesn’t apply to beta features like tracing. You can enable client-side PII redaction when intializing theAnalytics
class like so:
Error Handling
If an error occurs while sending events to Raindrop, an exception will be raised. Make sure to handle exceptions appropriately in your application.Configuration & helpers
- Debug logs –
debugLogs: true
prints every queued event. - Closing – call
await raindrop.close()
before your process exits to flush buffers.
AI Tracing (Beta)
AI tracing is currently in beta. We’d love your feedback while we continue to improve the experience!
- Visualize the full execution flow of your AI application
- Debug and optimize complex prompt chains
- Understand intermediate steps that led to a specific generated output
Enabling Tracing (Beta)
To keep bundle sizes small, tracing is disabled by default and requires extra steps to enable.raindrop-ai
package to serverExternalPackages
in your Next.js config.
Using withSpan
for Task Tracing (Beta)
The withSpan
method allows you to trace specific tasks or operations within your AI application. This is especially useful for tracking LLM requests. Any LLM call within the span will be automatically tracked, no further work required.
Parameters
Parameter | Type | Description |
---|---|---|
name | string | Name of the task for identification in traces |
properties | Record<string, string> (optional) | Key-value pairs for additional metadata |
inputParameters | unknown[] (optional) | Array of input parameters for the task |
Using withTool
for Tool Tracing (Beta)
The withTool
method allows you to trace any actions your agent takes. This could be as simple as saving or retrieving a memory, or using external services like web search or API calls. Tracing these actions helps you understand your agent’s behavior and what led up to the agent’s response.
Parameters
Parameter | Type | Description |
---|---|---|
name | string | Name of the tool for identification in traces |
version | number (optional) | Version number of the tool |
properties | Record<string, string> (optional) | Key-value pairs for additional metadata |
inputParameters | Record<string, any> (optional) | Record of input parameters for the tool |
traceContent | boolean (optional) | Flag to control whether content is traced |
suppressTracing | boolean (optional) | Flag to suppress tracing for this tool invocation |
Using with OTEL
If you have already set up OTEL tracing, using raindrop will conflict with it. To disable raindrop’s OTEL tracing, you can set thedisableTracing
option to true
when initializing the SDK.