Skip to main content
GET
Social media monitoring API for trending topics
Free - does not consume credits
Build a social media monitoring feed from ranked public topics. Radar covers Reddit, GitHub, technology news, search trends, Wikipedia, prediction markets, and startup growth records. Apply the relevant source, category, region, and time window to each request. Authenticate every Radar request. You need no active subscription. Radar consumes no credits.

What is a social media monitoring API?

A social media monitoring API turns public posts and trend signals into application-ready records. Radar exposes those records through one REST API endpoint. Each response uses structured JSON and stable field names. Radar gives every supported public source the same RadarItem shape. The GET /radar endpoint filters items by source, category, region, and age. The response ranks recent topics. Radar is not an exhaustive source archive. Radar combines several public topic sources. It does not mirror every social network. It also does not return X tweets, replies, followers, or timelines. For X-specific monitoring, Search Tweets finds brand mentions, hashtags, replies, and keyword matches. Use stable fields to compare titles, regions, timestamps, and scores. Avoid building separate source adapters. Compare trends across sources with Radar. Keep full social listening data in a dedicated platform API. Radar is not a social listening API for every major social platform. It does not replace platform APIs for unsupported social platforms. This API access returns ranked topics, not complete post, comment, or brand-mention records. Radar provides no keyword search volume or long-term history. Radar includes no more than 72 hours of history. Use Radar as a trending topics API for dashboards, alerts, and research queues. Start with a narrow time window. Apply source and category together or separately.
  1. Call GET /radar with your API key.
  2. Filter one source when the request names a platform.
  3. Limit the region when location affects the result.
  4. Sort or filter records by score in your application.
  5. Store each id to prevent duplicate dashboard rows.
  6. Continue with nextCursor while hasMore stays true.
Use source=reddit for Reddit posts. Use source=github for trending repositories. Other source identifiers follow the same pattern. This API access keeps one integration path across every supported feed. Radar returns recently indexed topics. It does not stream every new event. Use publishedAt for source timing. Use createdAt for indexing timing. Start each scheduled poll without after. Use nextCursor only for later pages within that poll. The contract guarantees neither cross-poll cursor lifetime nor snapshots. Deduplicate records by items[].id. Never use an item ID to order separate polling runs. Respect every 429 rate limit response. Wait before the next poll. This pattern tracks recent changes without promising a delivery deadline.

Can Radar track brand mentions and sentiment?

Radar has no keyword query. Radar does not provide a complete brand monitoring index. Compare returned titles and descriptions against a saved brand dictionary. That comparison covers only the current Radar feed. Radar does not analyze sentiment. The score value measures relevance, not sentiment. Do not classify a high score as positive or negative sentiment. Use Search Tweets to query exact X mentions. Use Radar for ranked cross-source topics. This split prevents false assumptions about brand-mention coverage.

How do you evaluate Radar topic accuracy?

Evaluate each item against the source fields. Do not treat a high trend score as proof that a claim is correct or popular everywhere.
  1. Match source and sourceId to the intended feed.
  2. Open url when present. Compare its title and description with the item.
  3. Compare publishedAt with createdAt to separate publication from indexing.
  4. Confirm region, category, and language match the intended alert.
  5. Interpret metadata with the source-specific fields documented below.
  6. Deduplicate by id, then retain the score used for that polling run.
Radar does not guarantee complete source coverage or independent fact verification. Review the original source before publishing an alert, report, or customer-facing dashboard entry.

Integrate Radar into a monitoring dashboard

Keep the dashboard pipeline simple. Request one page, validate the response, and render each RadarItem. Save the item ID, source, source ID, score, region, and timestamps. Group rows by source for a multi-feed dashboard. Rank dashboard rows by the returned score value, then filter them by publishedAt. Open the returned url when present. Use hasMore and nextCursor for background ingestion. Never construct a cursor. Keep it with matching filters during one pagination run. Changing source, category, region, or hours starts a different result set. Choose a searchable news API for exact keyword or domain matching. Use Radar for normalized records from its documented sources. This prevents claims about unsupported source coverage.

Headers

string
Your API key. Xquik also accepts session cookie authentication. Generate a key from the dashboard.

Query parameters

string
Filter with one source identifier. Use reddit for Reddit, github for GitHub, trustmrr for startup growth, hacker_news for technology news, google_trends for search trends, wikipedia for Wikipedia, or polymarket for prediction markets. Omit this filter to return every available stream.
string
Filter by category. One of: general, tech, dev, science, culture, politics, business, entertainment. Omit to return all categories. Xquik assigns categories from each source and item.
integer
default:"6"
Look-back window in hours. Range: 1-72. Smaller values return the most recent trends; larger values give a broader snapshot. Default is 6 hours.
integer
default:"50"
Items per page. Range: 1-100. Use smaller values (10-20) for quick polling, larger values for batch processing. Default is 50.
string
default:"global"
Region filter. Values: US, GB, TR, ES, DE, FR, JP, IN, BR, CA, MX, global. Default global returns items from all regions.
string
Cursor for pagination. Pass nextCursor from the previous response. Treat cursors as opaque base64 strings. Do not construct them.

Response

RadarItem[]
required
Array of radar items sorted by score (descending).
boolean
required
Whether more items are available after this page.
string
Pagination cursor. Present only when hasMore is true. Pass as the after query parameter on the next request.

RadarItem fields

string
required
Radar item identifier.
string
required
Item title.
string
Item description. The API omits this field when the source provides no value.
string
Link to the original source. The API omits this field when no link exists.
string
Source image URL. Startup growth items return the logo here. The API omits this field when no image exists.
string
required
Radar source identifier. Reuse this value in source to request the same stream.
string
required
Unique identifier within the source.
string
required
One of: general, tech, dev, science, culture, politics, business, entertainment.
string
required
Region code (e.g. US, TR, global).
string
required
The language value uses a BCP-47 code such as en, tr, or ja. The value und means the source did not identify a language.
number
required
This 0-10,000 relevance score ranks trending items. Higher values indicate a stronger trend signal within the result set.
object
required
Source-specific fields appear here. See the source tables below.
string
required
This shows when the source published the item or Radar found it. The value follows ISO 8601.
string
required
This records when Radar indexed the item. The value follows ISO 8601.

Metadata

Use source to interpret metadata. Multiword fields use camelCase in the default REST response and snake_case through API MCP.

Reddit

Every Reddit item identifies author, subreddit, and sourceFormat. Current items combine public listing discovery with server-rendered post fields. They include available post content, media, and engagement metrics. Legacy rows can still identify sourceFormat as json or rss. Exact public upvote and downvote counts are unavailable. Treat estimatedUpvotes and estimatedDownvotes as estimates because Reddit may fuzz the public score and ratio. numberComments is a count. Comment bodies are not returned.

Startup growth

Startup growth items always include mrr, growthPercent, last30Days, total, customers, activeSubscriptions, and onSale. Radar orders these items by reported 30-day revenue growth. metadata.rank is the revenue rank reported by the source. It does not equal the result position. The top-level imageUrl contains the startup logo when available.

Other sources

Examples

Response.

Pagination

Radar uses cursor-based pagination. When hasMore is true, pass nextCursor as after to fetch the next page.
Next steps. Trends guide for content strategy workflows · Create Tweet to post about trending topics · Compose to generate AI-drafted tweets from radar items