AI Recommendations (ZeroClick)
Semantic article search against a Dappier data model, authenticated by
verifying the zc-signature HMAC over the raw request bytes. The same
orchestration as POST /app/v2/search runs behind it.
Only marketplace-enabled data models are reachable — ZeroClick traffic
carries no Dappier tenant, so a private data model resolves to 404.
On success, usage is settled with ZeroClick and reported via the zc-usage
response header. The request is not charged on the Dappier platform.
Authorizations
HMAC signature minted by ZeroClick over the raw request bytes, of the form
t=<unix-seconds>,kid=<key-id>,v1=<hex-hmac>. The canonical string signed is
t \n METHOD \n path+query \n sha256hex(body) \n zc-request-id \n zc-agent-id,
and v1 = hex(HMAC_SHA256(secret, canonical)). Verified against the seller's
signing secret keyed by kid, with a ±300s timestamp skew bound.
A missing or invalid signature is rejected with 401
invalid_zeroclick_signature; this is never failed open.
Headers
ZeroClick request id, set by ZeroClick when it proxies the request. It is part of the signed canonical string, so a request without it cannot verify.
"zcreq_1"
Identifier of the buying agent. Used to partition logs/analytics by agent; becomes the synthesized caller identity (zeroclick:<agent-id>). Treated as an empty string when absent, which the signer must match.
"agt_1"
Query Parameters
Data model ID, starting with dm_. Browse and find data models in the Dappier Marketplace.
Optional search version override, forwarded to the retrieval pipeline.
Body
Natural language query, keyword, or URL. If a URL is specified, our AI analyzes the page context, summarizes it, and provides semantic recommendations based on the content.
The number of articles to return.
The site domain where AI recommendations are being displayed. Example format: dappier.com.
The minimum number of articles from the ref domain specified above. The rest of the articles will come from other sites within the RAG model.
Optional search type forwarded to the retrieval pipeline.
The search algorithm for retrieving articles.
most_recent: Retrieves the most recent articles by publication date. semantic (default): Retrieves contextually relevant articles based on the query or the page content from the provided URL. most_recent_semantic: Performs semantic search and retrieves the most recent articles by publication date. trending: Performs a trending search and retrieves the most relevant articles to trending keywords in the past 24 hours.
most_recent, semantic, most_recent_semantic, trending Page number for pagination.
Number of results per page.

