Primary direct tools
Composed research and enrichment
The two composed research tools accept
estimate, preview, and execute
modes. Estimate mode returns a cost and latency range without provider work.
Large executions can return an async job instead of holding one request open.
Market and fit analysis
Async jobs
Poll using the server-provided interval. Completed results are available only
to the authenticated workspace that created the job.
Every other catalog capability remains available through progressive
discovery. This includes workforce, audiences, jobs, identity, bulk
operations, iOS and Android apps, rankings, SDKs, websites, web technologies,
search, and social intelligence.
Presentation tools
superdata_render_people_results, superdata_render_company_results,
superdata_render_jobs_results, and superdata_render_company_brief turn
normalized findings into interactive result cards. Presentation tools are
read-only and do not consume credits.
Progressive fallback
superdata_search_capabilities
Search the catalog by natural-language intent or exact category when the
appropriate direct tool is not already clear.
Inputs
name, title, category, and an
optional description.
superdata_get_capability
Inspect the contract for one exact capability name returned by
superdata_search_capabilities.
Inputs
superdata_call
Execute one capability selected through discovery. Do not invent operation
names, path parameters, or taxonomy identifiers.
Inputs
ok: true- The executed
operation - Upstream
statusCodeandcontentType - Parsed provider
data responseBytesand atruncatedindicatorcreditsConsumed: 1
ok: false and a public error with code,
message, and an optional upstream statusCode. Failed execution restores the
reserved credit.
Capability areas
The catalog covers company and people intelligence, LinkedIn, email, technology and SDK signals, mobile applications, search results, taxonomy, segments, identity, privacy, and account usage. Search the live catalog rather than relying on a static list because available capabilities can change. For taxonomy-backed operations, runsegments_search_features first and use
the returned identifiers exactly as provided.