Search news events
search_news_events⚠️ DISCOVERY ONLY — NOT a company lookup. This returns a global feed of the
most recent events matching categories + company_locations (deduped by
company domain, ~10 rows). It takes NO company / domain / name input and
CANNOT be scoped to a specific company or to a list of known companies.
Use it ONLY for find-companies discovery ("surface net-new companies that
just did X"). To research signals for a SPECIFIC named company or an
uploaded prospect/account list, this is the WRONG tool — use
tavily_search with a company-scoped query (e.g. " customer
success news 2026") instead. search_monitor_events(company_domain=...)
works too, but only for companies the user already monitors.
Use this when the user asks to DISCOVER companies by a signal that maps to one of the supported categories below. Only fall back to tavily_search if this returns no results or the signal isn't covered.
Costs 0.3 Sliq credits per event record PredictLeads returns, up to 10 records per location (so up to 3 credits per location). A location query that fails is free.
Each event's company is then checked against criteria, one company at a
time, reading its name and the event, and only the events at matching
companies come back in events; the rest come back in rejected as
{name, reason}. It adds a few seconds per ten events.
When this runs in an agent, every event company with a domain is saved to
the Output tab as a company row (deduped by domain) with its verdict,
carrying the event in data.signals; the rejected ones are hidden from
the list's default view.
Supported categories (pass one or more in a single call):
Expansion: expands_facilities, expands_offices_in, expands_offices_to,
opens_new_location, increases_headcount_by, attends_event
Investment: receives_financing, goes_public, invests_into,
invests_into_assets, has_earnings, has_revenue, has_valuation
Leadership: hires, leaves, promotes, retires_from
Contract: signs_new_client, loses_client
Cost cutting: decreases_headcount_by, closes_offices_in
New offering: launches, is_developing, integrates_with
Partnership: partners_with, ends_partnership_with
Acquisition: acquires, merges_with, sells_assets_to
Challenges: declares_bankruptcy, files_suit_against, has_issues_with
Recognition: receives_award, recognized_as
Relational: identified_as_competitor_of, spins_off_company, spins_off_division
Dict with events, the events at matching companies, each with its
company's verdict ('qualified', or 'unclear' when the event says too
little to judge) and verdict_reason, and a null verdict when the
check couldn't run; rejected, the companies the check turned down,
as {name, reason}; count, every event found, rejected ones included;
and created_identifiers, the domains of matching companies this call
newly added to the list. If every location-query failed with an
upstream error (timeout / 5xx) and nothing landed, also
predictleads_available: False plus a human-readable error — a source
outage, NOT a genuinely empty feed. Partial success (some locations
errored, others returned events) stays silent and reports only the
events.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| agent_id | No | Optional — a specific agent to save the companies into. Omit it to use the running agent, which is the usual case. | |
| criteria | Yes | What makes a company a match for this search, in plain words: the event it must show and the kind of company (e.g. "A B2B software company that just raised a round"). Each event's company is checked against it, and the check reads nothing else of your instructions, so make it no narrower than what this call searches for. A role that is part of the signal belongs here; who you'll contact at a matching company doesn't, since the check reads that as a rule about which companies match. | |
| list_name | No | Short kebab slug naming the Output-tab list bucket. Absent, companies land in the 'default' list. A company already on the list keeps its verdict, so to screen against new criteria, pass a new `list_name`. | |
| categories | Yes | One or more category names from the enumerated list above (the parameter is constrained to exactly those values). Map common intents to the right name: "funding" → receives_financing, "IPO" → goes_public, "acquisition" → acquires, "layoffs" → decreases_headcount_by. Pass all relevant categories in a single call — they are fetched together at no extra cost. | |
| on_behalf_of | No | An agency client's email, from list_agency_clients, to run this call in that client's account: their agents, settings, connected accounts, and credits. Omit it to act on your own account. | |
| lookback_days | No | Days of news history to scan (default 30). A news scan passes the window from its trigger prompt. | |
| company_locations | No | Filter by country or US state. Pass a list to cover multiple locations (e.g. ["United States", "Canada"]). At most 10 API requests are made across all locations. Omit to search globally. Prefer country-level values (e.g. "United States") over individual cities/states unless the ICP is tightly city-scoped. |