Lusha

WHAT/guardrails here; HOW in reference/endpoints.md + lib/lusha.py. OpenAPI cached at reference/docs/lusha/raw/v2_openapi.md.

Auth & config

• Base URL: https://api.lusha.com
• Auth: header api_key (Lusha → API settings). Env LUSHA_API_KEY; never hardcode.

Two modes

• Enrichment (known target): enrich_person_email / enrich_person_phone / enrich_company (single) or enrich_companies (bulk). Consumes credits on reveal, at the price of the field the method reveals.
• Prospecting (search → enrich): search_contacts/search_companies (cheap, returns requestId + ids) → enrich_contact_emails/enrich_contact_phones/enrich_companies (reveal the chosen ids, consumes credits).

Guardrails

• Credits: reveal/enrich consumes credits; search is cheap. A phone reveal costs 5x an email one, and the method fixes which you get — there's no field argument to get it wrong. Pilot one, confirm, then bulk. get_usage() is a safe read-only pilot.
• V1 person API is deprecated — this adapter uses v2 only.

Handoff

Enrichment/prospecting layer: search or take a shortlist → Lusha for verified emails/direct-dials → ATS (recruit-crm) / sequencer. Enrich late, on the filtered set.

Field notes (production experience)

Field-shape note: these are vendor-native operational notes. The client returns the raw vendor JSON and uses the method names in this file — read field shapes accordingly (no normalized-wrapper / result.data. prefix).

Strength: direct dials (mobile + desk), not just HQ. Coverage global, strongest NA + Europe. ~$0.05/credit.

• Person input: LinkedIn URL (best - must contain linkedin.com/in/; Sales Navigator/Recruiter URLs fail), OR email, OR first_name+last_name+(company_name or company_domain). Name+company is the minimum combo - don't pass company_name alone.
• Company enrich: domain/name/Lusha ID; don't include http:// or www. in domain values.
• Output: email at emails[].email (or top-level email); phone at phones[].number - check the array first, top-level phone isn't always populated.
• Fields: an *_email method reveals emails, a *_phone method phones; intent data comes from the *_signals methods.
• Search: filters seniority, companySize (e.g. "201-500"), department, etc.; returns { contacts, pagination:{ page, pageSize, total, totalPages } }.

Callable surface — lib/lusha.py

Import: from lib.lusha import Lusha → instantiate Lusha() (reads key from env). Base: https://api.lusha.com. Generic passthrough: request(method, path, *, params, json).

• company_filter_types() -> dict — GET /v3/companies/prospecting/filters — list company filter types.
• company_filter_values(filter_type: str, *, query: str | None = None) -> dict — GET /v3/companies/prospecting/filters/{type} — valid values (sizes/revenues/sics/naics/
• company_lookalikes(payload: dict) -> dict — POST /v3/companies/lookalike — AI lookalikes from 5–100 seed companies (domains/linkedin).
• company_signals(payload: dict) -> dict — POST /v3/companies/signals — hiring/headcount/web-traffic/IT-spend/news signals for up to
• contact_filter_types() -> dict — GET /v3/contacts/prospecting/filters — list contact filter types (+ whether query required).
• contact_filter_values(filter_type: str, *, query: str | None = None) -> dict — GET /v3/contacts/prospecting/filters/{type} — valid values for a contact filter
• contact_lookalikes(payload: dict) -> dict — POST /v3/contacts/lookalike — AI lookalikes from 5–100 seed contacts (linkedinUrl/email/
• contact_signals(payload: dict) -> dict — POST /v3/contacts/signals — job-change/promotion events for up to 100 contact ids
• enrich_companies(ids: list[str], **opts) -> dict — POST /v3/companies/enrich — reveal firmographics (size/revenue/industry/tech/funding/
• enrich_company(*, domain: str | None = None, name: str | None = None) -> dict — Convenience: search-and-enrich a SINGLE company by domain or name (no http/www in domain).
• enrich_contact_emails(ids: list[str]) -> dict — POST /v3/contacts/enrich — reveal the EMAILS of contacts already found via search. ids =
• enrich_contact_phones(ids: list[str]) -> dict — POST /v3/contacts/enrich — reveal the PHONES of contacts already found via search. ids =
• enrich_person_email(*, linkedin_url: str | None = None, email: str | None = None, first_name: str | None = None, last_name: str | None = None, company_name: str | None = None, company_domain: str | None = None) -> dict — Convenience: search-and-enrich the EMAIL of a SINGLE contact (wraps
• enrich_person_phone(*, linkedin_url: str | None = None, email: str | None = None, first_name: str | None = None, last_name: str | None = None, company_name: str | None = None, company_domain: str | None = None) -> dict — Convenience: search-and-enrich the PHONE of a SINGLE contact (wraps
• get_usage() -> dict — GET /v3/account/usage — credits (total/used/remaining), rate limits, plan, and per-action
• prospecting_companies(payload: dict) -> dict — POST /v3/companies/prospecting — filter-based company search (size/revenue/industry/tech/
• prospecting_contacts(payload: dict) -> dict — POST /v3/contacts/prospecting — filter-based ICP search (title/seniority/department/
• search_and_enrich_companies(companies: list[dict], **opts) -> dict — POST /v3/companies/search-and-enrich — find + reveal companies in one call (id/name/domain).
• search_and_enrich_contact_emails(contacts: list[dict], *, options: dict | None = None) -> dict — POST /v3/contacts/search-and-enrich — find + reveal EMAILS in ONE call. contacts:
• search_and_enrich_contact_phones(contacts: list[dict], *, options: dict | None = None) -> dict — POST /v3/contacts/search-and-enrich — find + reveal PHONES in ONE call. Same contacts
• search_companies(companies: list[dict], **opts) -> dict — POST /v3/companies/search — preview companies by id/name/domain (max 100). Returns has
• search_contacts(contacts: list[dict], *, filters: dict | None = None, **opts) -> dict — POST /v3/contacts/search — preview contacts by identifier (id / linkedinUrl / email /