Get an Observation
Retrieves a single observation based on the provided observation ID.
Subscription Endpoint
This endpoint has a special subscription requirement. Your API Key needs to be subscribed to this endpoint in order to access it; otherwise it will return a 403 Forbidden error. Please contact Forestreet support to get subscribed.
An Observation is a piece of information or data point that has been collected and stored in our system. It serves as the foundational unit of information to join together as Events, and can represent various types of data such as news articles, social media posts, analyst reports, or any other relevant information that we want to track and analyze. Each observation contains metadata about its source, content, publication time, and other relevant attributes that allow us to process and analyze it effectively.
Authorizations
Pass a static API key for every request, provided by your customer support.
Path Parameters
The unique identifier of the observation
"69e89d74df2753cf786ef002"
Response
The response containing pagination metadata, and an array of zero or more events that satisfy the query parameters.
The response schema for the Get Observation endpoint.
Unique identifier for the observation
"69e89d74df2753cf786ef002"
A deduplication identifier for the observation, used to identify and merge duplicate observations.
"NewsArticle::url(https://www.dw.com/en/czechias-lithium-mine-plans-spark-resistance/video-76878114)"
The service from which the observation was sourced, e.g. 'News Monitor'.
"News Monitor"
Title of the observation. For news articles, this will typically be the native title of the article from the publisher.
"Czechia's lithium mine plans spark resistance"
An array of text snippets that provide the content of the observation. For news articles, this may include the article summary, key sentences, or other relevant text extracted from the article. If an article failed to scrape, it may only contain 1 content item which is the search engine snippet.
The content items may have been pruned before being stored on our system, so that only the most relevant information is retained. It is not guaranteed to be the full text of the original source.
An array of numbers representing the embedding vector for the observation's content.
A dictionary of weights for the observation, where the keys are weight names (e.g. 'Media Bias Multiplier') and the values are the corresponding weight values. These weights can be used to adjust the importance of the observation in various algorithms, such as event clustering or impact assessment.
The date and time when the observation was published or recorded. These are always UTC.
"2026-04-21T17:02:22.885Z"
The date and time when the observation was added to our system. These are always UTC.
"2026-04-22T10:02:22.885Z"
The cluster label assigned to this observation, if it has been clustered.
URL of the observation, if it was sourced and scraped from the internet. Only available on publicly sourced observations such as News Articles.
"https://www.dw.com/en/czechias-lithium-mine-plans-spark-resistance/video-76878114"