Command Reference

Every discolike command, grouped the way the CLI groups them. Each entry lists the exact usage line, every option, and a runnable example.

New here? Start with Getting Started for install, authentication, and output formats.

account

Account usage and quota.

account usage

Month-to-date usage: requests, records, and spend.

discolike account usage [OPTIONS]

Flag	Type	Default	Description
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike account usage

{
  "requests_mtd": null,
  "records_mtd": null,
  "spend_mtd": null,
  "account_status": "active",
  "month_to_date_requests": 14,
  "month_to_date_records": 11794,
  "month_to_date_spend": 24.99,
  "max_spend": "unlimited",
  "total_available_spend": "unlimited",
  "carryover_credits": 0.0,
  "top_up_credits": 0.0,
  "usage_summary": {
    "2026-08": [
      {
        "access_id": "••••••••isco",
        "description": "[API] Internal key",
        "requests": 13,
        "total_records": 11744,
        "monthly_spend": 24.79
      }
    ],
    ...
  },
  ...
}

REST endpoint: Usage API. SDK equivalent: client.account.usage().

append

Enrich a CSV of domains with DiscoLike datasets. This command is synchronous and returns the enriched rows directly.

discolike append [OPTIONS] FILE

Flag	Type	Default	Description
FILE (argument)	path	required	CSV of domains to enrich.
--dataset	str	required	Dataset to append (repeatable): bizdata, redirects, domain_status, growth, vendors. Omitting it exits 2 before any request.
--domain-column	str	none	Column in the CSV that holds domains.
--csv / --no-csv	bool	unset	Request the enriched rows as CSV (written via --output).
--output	path	none	File to write a CSV response to.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--query-id	str	none	Saved query ID whose domains are included alongside the file/—domain ones (repeatable).

Pass --output whenever the API returns CSV bytes; without it the command exits 2 with --output is required when the response is CSV bytes. With --output it writes the bytes verbatim and prints {"written": "<path>", "bytes": <count>}.

discolike append domains.csv --dataset bizdata --dataset growth

discolike append domains.csv --domain-column website --csv --output enriched.csv

REST endpoint: Append API. SDK equivalent: client.append().

auth

Manage credentials: log in (browser or API key), check status, log out.

auth login and auth status honour the root --base-url: auth login logs in to that host’s authorization server and auth status verifies against it. A root --api-key passed explicitly on the command line switches auth login to the API-key flow with that key, and is the key auth status reports as "source": "option". Credential resolution, consent reuse, and the config file are covered in Getting Started.

auth login

Log in via the browser (OAuth) or with an API key, verify, and save to the local config file.

discolike auth login [OPTIONS]

Flag	Type	Default	Description
--api-key	str	none	Log in with an API key instead of the browser. Create one at https://app.discolike.com/account/management/keys.
--method	str	oauth	oauth (browser login, default) or api_key (prompts for a key unless --api-key is given). Any other value is a usage error, exit 2.
--no-browser	flag	off	Print the login URL instead of opening a browser.
--port	int	0	Fixed loopback port for the browser redirect (default: random; use with SSH).

Browser flow (the default). The CLI fetches the authorization server’s metadata from the API host, registers itself as a public PKCE client or reuses the registration saved from an earlier login, starts a listener on http://127.0.0.1:<port>/callback, and prints the authorization URL to stderr before opening it. If no browser can be opened it says so and leaves the URL for you to open by hand. It then waits up to 180 seconds for the redirect, exchanges the code for tokens, verifies the session with a usage call, and saves it. On success the payload goes to stderr, not stdout:

{"logged_in": true, "method": "oauth", "expires_at": "2026-08-28T20:15:00+00:00"}

discolike auth login

On a remote machine, pin the port, forward it, and open the printed URL in a local browser:

ssh -L 51234:127.0.0.1:51234 build-host
discolike auth login --no-browser --port 51234

API-key flow. --api-key KEY uses that key; --method api_key prompts for one with hidden input. The key is verified with a usage call before anything is written, and a bad key exits 3 and writes nothing. The success payload is {"logged_in": true, "source": "api_key"} on stderr.

discolike auth login --api-key dk-live-...

discolike auth login --method api_key

Failures of the browser flow itself exit 1 with a single JSON line on stderr:

message	Cause
Timed out after 180s waiting for the browser login	No redirect arrived within the wait
Authorization failed: <error_description or error>	The authorization server redirected back with an error, such as access_denied when consent was refused
Invalid OAuth callback (state mismatch or missing code)	The redirect did not carry the expected state and code
Token exchange failed: <error>	The code could not be exchanged for tokens

{"error": "LoginError", "message": "Authorization failed: access_denied"}

A saved client registration the server no longer accepts (invalid_client or unauthorized_client) is discarded and replaced automatically, with the browser flow run once more; only a second rejection surfaces as a LoginError. Errors from the authorization server outside the browser round trip, such as unreachable metadata or a rejected registration, are AuthenticationError and exit 3 like any other API error.

auth status

Show which credential is in use (option, env, or config) and verify it against the API. No options.

discolike auth status

Field	Present	Value
source	always	option for a root --api-key, env for DISCOLIKE_API_KEY, config for the config file
method	always	oauth or api_key
api_key	API key only	The key masked to its last four characters
expires_at	OAuth only	Access-token expiry as an ISO 8601 UTC timestamp
expired	OAuth only	Whether the access token is past expires_at. The session is still refreshed on the next command, so this can be true while valid is true
valid	always	true; an invalid credential raises instead

The option is checked first, then the environment variable, then the config file. With no credential anywhere, it exits 3.

discolike auth status

{
  "source": "config",
  "method": "oauth",
  "expires_at": "2026-08-28T20:15:00+00:00",
  "expired": false,
  "valid": true
}

{
  "source": "env",
  "method": "api_key",
  "api_key": "…isco",
  "valid": true
}

auth logout

Delete saved credentials from the local config file (the registered OAuth client is kept). No options. Idempotent - logging out twice is not an error.

discolike auth logout

discolike auth logout

Prints {"logged_out": true}. The OAuth session or API key is removed; the oauth_client registration stays so the next browser login does not ask for consent again.

bulk

Volume pulls with checkpoint and resume: walk the company index past the 10,000-per-search ceiling, size a contact pull for free, then pull N contacts per company into one CSV. Filters come from --params-file, --param, and the common flags; the paging parameters are managed by the command and a supplied value is dropped with a note on stderr.

Every bulk command validates the request against the SDK model before the first billable call, keeps one request in flight under --rate-limit, retries on 429 and 5xx above the SDK’s own retries, prints progress to stderr, and prints one JSON summary to stdout. Rate limits are per plan: /discover and /contacts allow 5 per minute on Starter, 10 on Pro, 15 on Team, 25 on Company, and 50 on Enterprise; /contacts/count allows six times that.

bulk companies

Walk the company index past 10,000 results, one exclusion list per page, appending to a CSV.

discolike bulk companies [OPTIONS]

Flag	Type	Default	Description
--icp-prompt	str	none	Natural-language ideal customer profile description.
--domain	str	none	Seed domain for lookalike matching (repeatable).
--phrase-match	str	none	Phrase the company website must contain (repeatable).
--negate-phrase-match	str	none	Negate the --phrase-match filter (repeatable).
--category	str	none	Industry category filter (repeatable).
--country	str	none	ISO country code filter (repeatable).
--state	str	none	State or region filter (repeatable).
--employee-range	str	none	Employee count range filter.
--revenue-range	str	none	Revenue range filter.
--variance	str	none	Result diversity: LOW, MID_LOW, MEDIUM, MID_HIGH, HIGH, UNRESTRICTED.
--exclusion-query-id	str	none	Saved query ID whose results are excluded (repeatable). Also the way to reuse the lists a previous run created.
--param	str	none	Extra API parameter as KEY=VALUE (comma-separates into a list); see the endpoint reference (repeatable).
--params-file	path	none	JSON object of API parameter names (an app form copied over); --param and named flags override its values.
--max-companies	int	10000	Stop once this many companies are in --out.
--page-size	int	10000	Companies per discover call (20 to 10000).
--run-name	str	bulk	Prefix for the exclusion lists created per page (<run-name>-round-N).
--out	path	companies.csv	CSV to append each page to; resumes if present.
--overwrite	flag	off	Ignore an existing --out and start clean instead of resuming.
--rate-limit	int	10	Calls per minute to stay under on /discover.

Mechanics: each page is a discover call at --page-size; its net-new domains are appended to the CSV (domain, name, country, employees, similarity) and saved as an exclusion list named <run-name>-round-N, which is passed as exclusion_query_id on the next page (a tail shorter than 20 domains rides on exclude_domain instead, since saved lists need at least 20). The loop stops when a page returns nothing net-new, when a page comes back shorter than requested, or at --max-companies. Rerunning with the same --out reads the CSV back and rebuilds the exclusion set in 10,000-domain lists before continuing; pass the original ids as --exclusion-query-id to reuse them instead. Reach is bounded by the account’s exclusion capacity of 250,000 domains; the command warns there, and the fix is to slice the ICP (by state, employee band, sub-vertical) into separate runs.

max_records, offset, exclude_domain, and exclusion_query_id inside --params-file or --param are managed by the command and dropped with a note.

discolike bulk companies --params-file form.json --max-companies 50000 \
  --run-name agencies --out companies.csv

round 1: +10,000 net-new (10,000 total)
round 2: +10,000 net-new (20,000 total)
...

{
  "companies": 43210,
  "new": 43210,
  "rounds": 5,
  "exclusion_query_ids": ["…", "…", "…", "…"],
  "out": "companies.csv"
}

REST endpoint: Discover API and Queries API. SDK equivalent: client.discover() in a loop with client.queries.create_exclusion_list().

bulk estimate

Size a contact pull for free: matching contacts across the domain list, and the cap at --per-company.

discolike bulk estimate [OPTIONS]

Flag	Type	Default	Description
--domains-file	path	required	CSV with a domain column, or one domain per line.
--per-company	int	10	Contacts per company the pull would ask for (1 to 100).
--icp-prompt	str	none	Natural-language ICP prompt used to derive contact filters.
--summary	str	none	Filter by profile summary text (semantic search); ranks who comes back per company.
--negate-summary	str	none	Exclude contacts matching this summary description.
--seniority	str	none	Filter by seniority level (repeatable).
--negate-seniority	str	none	Exclude seniority levels (repeatable).
--department	str	none	Filter by department (repeatable).
--negate-department	str	none	Exclude departments (repeatable).
--title	str	none	Filter by job title (repeatable).
--negate-title	str	none	Exclude job titles (repeatable).
--person-country	str	none	Filter by contact country (repeatable).
--person-state	str	none	Filter by contact state/region (repeatable).
--has-email / --no-has-email	bool	on	Only contacts with an email address.
--exclusion-query-id	str	none	Saved query ID whose results are excluded (repeatable).
--param	str	none	Extra API parameter as KEY=VALUE (comma-separates into a list); see the endpoint reference (repeatable).
--params-file	path	none	JSON object of API parameter names; --param and named flags override its values.
--rate-limit	int	30	Calls per minute on the free /contacts/count.

Runs contacts count over the list in 1,000-domain slices with the same filters bulk contacts will use and sums the result. contacts_capped is the most the pull can return at --per-company; the real yield is usually lower, two to three usable emails per company is a normal outcome at 5 to 10 per company once ranking and --negate-summary apply.

discolike bulk estimate --domains-file companies.csv --per-company 10 \
  --summary "growth marketing lead" --negate-summary "controller, accountant, bookkeeping"

{
  "companies": 43210,
  "contacts_available": 118004,
  "contacts_capped": 118004,
  "per_company": 10
}

REST endpoint: Contacts API. SDK: client.contacts.count().

bulk contacts

Pull up to --per-company contacts for every domain in the file into one CSV, resumable by slice.

discolike bulk contacts [OPTIONS]

Uses the same filters as bulk estimate, with output and resume controls:

Flag	Type	Default	Description
--domains-file	path	required	CSV with a domain column, or one domain per line.
--per-company	int	10	Contacts per company the pull would ask for (1 to 100).
--icp-prompt	str	none	Natural-language ICP prompt used to derive contact filters.
--summary	str	none	Filter by profile summary text (semantic search); ranks who comes back per company.
--negate-summary	str	none	Exclude contacts matching this summary description.
--seniority	str	none	Filter by seniority level (repeatable).
--negate-seniority	str	none	Exclude seniority levels (repeatable).
--department	str	none	Filter by department (repeatable).
--negate-department	str	none	Exclude departments (repeatable).
--title	str	none	Filter by job title (repeatable).
--negate-title	str	none	Exclude job titles (repeatable).
--person-country	str	none	Filter by contact country (repeatable).
--person-state	str	none	Filter by contact state/region (repeatable).
--has-email / --no-has-email	bool	on	Only contacts with an email address.
--exclusion-query-id	str	none	Saved query ID whose results are excluded (repeatable).
--param	str	none	Extra API parameter as KEY=VALUE (comma-separates into a list); see the endpoint reference (repeatable).
--params-file	path	none	JSON object of API parameter names; --param and named flags override its values.
--out	path	contacts.csv	CSV to append to; <out>.checkpoint tracks finished slices.
--overwrite	flag	off	Ignore an existing --out and its checkpoint and start clean.
--rate-limit	int	10	Calls per minute to stay under on /contacts/discover.

Mechanics: /contacts/discover ignores offset when results_by_company is set, so the command paginates by slicing the domain list at 10000 / per-company domains per call, with results_by_company at --per-company and max_records at the slice’s cap. Each response is flattened to one row per contact (persona_id, first_name, last_name, name, title, seniority, department, email, email_validated, phone, linkedin, connections, domain, company_name, country, state, industry, employees, revenue_range, jobstart_date), de-duplicated on persona_id against the run and the rows already in the CSV, and appended; the slice index is then written to <out>.checkpoint. A rerun skips finished slices. Billing is net-new only with a 90-day cache, so a rerun after a crash does not re-charge for records already pulled.

domain, max_records, offset, results_by_company, and max_companies inside --params-file or --param are managed by the command and dropped with a note.

discolike bulk contacts --domains-file companies.csv --per-company 10 \
  --summary "growth marketing lead" --negate-summary "controller, accountant, bookkeeping" \
  --out contacts.csv

43,210 companies | 44 slice(s) of 1000 | 44 to run
  slice 1/44: 2,391 contacts (2,391 this run)
  slice 2/44: 2,402 contacts (4,793 this run)
...

{
  "companies": 43210,
  "slices": 44,
  "slices_run": 44,
  "contacts": 104388,
  "out": "contacts.csv",
  "checkpoint": "contacts.csv.checkpoint"
}

REST endpoint: Contacts API. SDK: client.contacts.discover() per slice.

company

Company profiles by domain: firmographics, scores, growth, redirects, vendors, subsidiaries, and public links. Every subcommand takes a required DOMAIN argument and is synchronous.

SDK equivalents for this whole group live on client.companies.

company data

Full company profile (firmographics) for a domain.

discolike company data [OPTIONS] DOMAIN

Flag	Type	Default	Description
DOMAIN (argument)	str	required	Company domain, e.g. stripe.com.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike company data stripe.com

{
  "domain": "stripe.com",
  "name": "Stripe",
  "status": {
    "status": "active",
    "confidence": 0.89
  },
  "score": 701,
  "start_date": "2011-01-04",
  "end_date": null,
  "address": {
    "street": "354 Oyster Point Blvd",
    "city": "South San Francisco",
    "state": "California",
    "zip": "94080",
    "country": "US"
  },
  "phones": [
    "+16504279276",
    "+18889262289"
  ],
  ...
}

REST endpoint: BizData API. SDK: client.companies.data().

company score

Company score for a domain.

discolike company score [OPTIONS] DOMAIN

Flag	Type	Default	Description
DOMAIN (argument)	str	required	Company domain, e.g. stripe.com.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike company score stripe.com

{
  "domain": "stripe.com",
  "score": 701,
  "parameters": {
    "base_score": 688.0,
    "recency_multiplier": 1.0,
    "growth_boost": 12.06,
    "lookback_360": 476,
    "lookback_720": 291,
    "expiration_penalty": 1.0
  },
  "first_event": "2011-01-04"
}

REST endpoint: Score API. SDK: client.companies.score().

company growth

Growth signals for a domain.

discolike company growth [OPTIONS] DOMAIN

Flag	Type	Default	Description
DOMAIN (argument)	str	required	Company domain, e.g. stripe.com.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike company growth stripe.com

REST endpoint: Growth API. SDK: client.companies.growth().

company redirects

Domain redirects for a company domain.

discolike company redirects [OPTIONS] DOMAIN

Flag	Type	Default	Description
DOMAIN (argument)	str	required	Company domain, e.g. stripe.com.
--match	str	none	Domain match mode, e.g. loose.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike company redirects stripe.com --match loose

REST endpoint: Redirects API. SDK: client.companies.redirects().

company vendors

Vendors associated with a company domain.

discolike company vendors [OPTIONS] DOMAIN

Flag	Type	Default	Description
DOMAIN (argument)	str	required	Company domain, e.g. stripe.com.
--match	str	none	Domain match mode, e.g. loose.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike company vendors stripe.com

REST endpoint: Vendors API. SDK: client.companies.vendors().

company subsidiaries

Subsidiaries of a company domain.

discolike company subsidiaries [OPTIONS] DOMAIN

Flag	Type	Default	Description
DOMAIN (argument)	str	required	Company domain, e.g. stripe.com.
--match	str	none	Domain match mode, e.g. loose.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike company subsidiaries stripe.com --match loose

REST endpoint: Subsidiaries API. SDK: client.companies.subsidiaries().

company public-links

Public profile links for a domain from a given source. --source is required.

discolike company public-links [OPTIONS] DOMAIN

Flag	Type	Default	Description
DOMAIN (argument)	str	required	Company domain, e.g. stripe.com.
--source	str	required	Public link source, e.g. crunchbase.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike company public-links stripe.com --source crunchbase

Omitting --source is a usage error and exits 2.

REST endpoint: PublicLink API. SDK: client.companies.public_links().

contacts

Find contacts: search and count by filters, look up or match individuals, and run bulk/generative discovery jobs. SDK equivalents live on client.contacts.

search, count and discover share one filter block, repeated in full in each table below.

contacts search

Search contacts matching the given filters.

discolike contacts search [OPTIONS]

Flag	Type	Default	Description
--icp-prompt	str	none	Natural-language ICP prompt used to derive contact filters.
--seniority	str	none	Filter by seniority level (repeatable).
--negate-seniority	str	none	Exclude seniority levels (repeatable).
--department	str	none	Filter by department (repeatable).
--negate-department	str	none	Exclude departments (repeatable).
--title	str	none	Filter by job title (repeatable).
--negate-title	str	none	Exclude job titles (repeatable).
--domain	str	none	Filter by company domain (repeatable).
--domains-file	path	none	CSV with a domain column, or one domain per line; merged with --domain.
--person-country	str	none	Filter by contact country (repeatable).
--negate-person-country	str	none	Exclude contact countries (repeatable).
--filter-industry	str	none	Filter by company industry (repeatable).
--negate-filter-industry	str	none	Exclude company industries (repeatable).
--filter-country	str	none	Filter by company country (repeatable).
--negate-filter-country	str	none	Exclude company countries (repeatable).
--employee-range	str	none	Company employee range, e.g. 50-200.
--has-email / --no-has-email	bool	unset	Only contacts with (or without) an email address.
--jobstart-date	str	none	Job start date filter: min date or ‘min,max’ range, e.g. 2025-01-01 or 2025-01-01,2025-06-30.
--max-records	int	none	Maximum number of contacts to return.
--offset	int	none	Number of records to skip for pagination.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--param	str	none	Extra key=value query parameter forwarded to the SDK (repeatable).
--name	str	none	Filter by contact name (partial match supported).
--summary	str	none	Filter by profile summary text (semantic search).
--negate-summary	str	none	Exclude contacts matching this summary description.
--skills	str	none	Filter by skill (repeatable).
--person-state	str	none	Filter by contact state/region (repeatable).
--persona-id	int	none	Filter by persona ID (repeatable).
--filter-state	str	none	Filter by company state/region (repeatable).
--negate-filter-state	str	none	Exclude contacts at companies in this state (repeatable).
--email-validated	bool	none	Only contacts with (or without) a validated email address.
--has-phone	bool	none	Only contacts with (or without) a phone number.
--has-mobile	bool	none	Only contacts with (or without) a mobile phone number.
--has-linkedin	bool	none	Only contacts with (or without) a LinkedIn profile.
--min-connections	int	none	Minimum LinkedIn connections required.
--inclusion-query-id	str	none	Include only contacts from companies in this saved query (repeatable).
--exclusion-query-id	str	none	Exclude contacts from companies in this saved query (repeatable).
--max-companies	int	none	Maximum number of enriched companies to return; cannot be combined with —max-records.
--results-by-company	int	none	Maximum contacts per company domain (default 5; 0 removes the cap).
--include-search-contacts	bool	none	Include contacts from the search index (broader coverage).
--consensus	int	none	Number of query vectors to combine for consensus search.
--params-file	path	none	JSON object of API parameter names (an app form copied over); --param and named flags override its values.

discolike contacts search --seniority vp --department marketing \
  --domain stripe.com --has-email --max-records 5

--icp-text is not accepted here; it exists only on contacts generate.

REST endpoint: Contacts API. SDK: client.contacts.search().

contacts count

Count contacts matching the given filters. Same filter block as search, without --max-records and --offset.

discolike contacts count [OPTIONS]

Flag	Type	Default	Description
--icp-prompt	str	none	Natural-language ICP prompt used to derive contact filters.
--seniority	str	none	Filter by seniority level (repeatable).
--negate-seniority	str	none	Exclude seniority levels (repeatable).
--department	str	none	Filter by department (repeatable).
--negate-department	str	none	Exclude departments (repeatable).
--title	str	none	Filter by job title (repeatable).
--negate-title	str	none	Exclude job titles (repeatable).
--domain	str	none	Filter by company domain (repeatable).
--domains-file	path	none	CSV with a domain column, or one domain per line; merged with --domain.
--person-country	str	none	Filter by contact country (repeatable).
--negate-person-country	str	none	Exclude contact countries (repeatable).
--filter-industry	str	none	Filter by company industry (repeatable).
--negate-filter-industry	str	none	Exclude company industries (repeatable).
--filter-country	str	none	Filter by company country (repeatable).
--negate-filter-country	str	none	Exclude company countries (repeatable).
--employee-range	str	none	Company employee range, e.g. 50-200.
--has-email / --no-has-email	bool	unset	Only contacts with (or without) an email address.
--jobstart-date	str	none	Job start date filter: min date or ‘min,max’ range, e.g. 2025-01-01 or 2025-01-01,2025-06-30.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--param	str	none	Extra key=value query parameter forwarded to the SDK (repeatable).
--name	str	none	Filter by contact name (partial match supported).
--summary	str	none	Filter by profile summary text (semantic search).
--negate-summary	str	none	Exclude contacts matching this summary description.
--skills	str	none	Filter by skill (repeatable).
--person-state	str	none	Filter by contact state/region (repeatable).
--persona-id	int	none	Filter by persona ID (repeatable).
--filter-state	str	none	Filter by company state/region (repeatable).
--negate-filter-state	str	none	Exclude contacts at companies in this state (repeatable).
--email-validated	bool	none	Only contacts with (or without) a validated email address.
--has-phone	bool	none	Only contacts with (or without) a phone number.
--has-mobile	bool	none	Only contacts with (or without) a mobile phone number.
--has-linkedin	bool	none	Only contacts with (or without) a LinkedIn profile.
--min-connections	int	none	Minimum LinkedIn connections required.
--inclusion-query-id	str	none	Include only contacts from companies in this saved query (repeatable).
--exclusion-query-id	str	none	Exclude contacts from companies in this saved query (repeatable).
--max-records	int	none	Maximum number of contacts to count (20-10000).
--max-companies	int	none	Maximum number of enriched companies to return; cannot be combined with —max-records.
--offset	int	none	Number of records to skip for pagination.
--results-by-company	int	none	Maximum contacts per company domain (default 5; 0 removes the cap).
--include-search-contacts	bool	none	Include contacts from the search index (broader coverage).
--consensus	int	none	Number of query vectors to combine for consensus search.
--params-file	path	none	JSON object of API parameter names (an app form copied over); --param and named flags override its values.

discolike contacts count --seniority vp --domain stripe.com

{
  "count": 56
}

REST endpoint: Contacts API. SDK: client.contacts.count().

contacts lookup

Look up a single contact by persona ID, LinkedIn URL, or email.

discolike contacts lookup [OPTIONS]

Flag	Type	Default	Description
--persona-id	int	none	Look up by persona ID.
--linkedin	str	none	Look up by LinkedIn profile URL.
--email	str	none	Look up by email address.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

All three identifiers are optional, and you can pass several at once - the API decides which to use.

discolike contacts lookup --email jane@example.com

REST endpoint: Contacts API. SDK: client.contacts.lookup().

contacts match

Match a person name to contact records.

discolike contacts match [OPTIONS] NAME

Flag	Type	Default	Description
NAME (argument)	str	required	Full name of the person to match.
--company-name	str	none	Company name to narrow the match.
--domain	str	none	Company domain to narrow the match.
--person-country	str	none	Contact country to narrow the match.
--limit	int	none	Maximum number of matches to return.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike contacts match "Jane Doe" --company-name "Acme Corp" --domain acme.com --limit 5

REST endpoint: Contact Match API. SDK: client.contacts.match().

contacts bulk-match

Match many person queries in one async job. Task family contactmatch.

discolike contacts bulk-match [OPTIONS]

Flag	Type	Default	Description
--queries-file	path	required	Path to a JSON file with an array of match query objects.
--enrich / --no-enrich	bool	unset	Enable or disable enrichment of matched contacts.
--limit	int	none	Maximum number of matches per query.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

The file must parse as JSON and contain a top-level array; anything else exits 2.

discolike contacts bulk-match --queries-file queries.json --enrich --limit 5 --wait

Poll a non-waiting run with discolike discogen status <task_id> --family contactmatch.

REST endpoint: Contact Bulk Match API. SDK: client.contacts.bulk_match().

contacts discover

Discover contacts grouped by company for the given filters.

discolike contacts discover [OPTIONS]

Flag	Type	Default	Description
--icp-prompt	str	none	Natural-language ICP prompt used to derive contact filters.
--seniority	str	none	Filter by seniority level (repeatable).
--negate-seniority	str	none	Exclude seniority levels (repeatable).
--department	str	none	Filter by department (repeatable).
--negate-department	str	none	Exclude departments (repeatable).
--title	str	none	Filter by job title (repeatable).
--negate-title	str	none	Exclude job titles (repeatable).
--domain	str	none	Filter by company domain (repeatable).
--domains-file	path	none	CSV with a domain column, or one domain per line; merged with --domain.
--person-country	str	none	Filter by contact country (repeatable).
--negate-person-country	str	none	Exclude contact countries (repeatable).
--filter-industry	str	none	Filter by company industry (repeatable).
--negate-filter-industry	str	none	Exclude company industries (repeatable).
--filter-country	str	none	Filter by company country (repeatable).
--negate-filter-country	str	none	Exclude company countries (repeatable).
--employee-range	str	none	Company employee range, e.g. 50-200.
--has-email / --no-has-email	bool	unset	Only contacts with (or without) an email address.
--jobstart-date	str	none	Job start date filter: min date or ‘min,max’ range, e.g. 2025-01-01 or 2025-01-01,2025-06-30.
--max-records	int	none	Maximum number of contacts to return.
--offset	int	none	Number of records to skip for pagination.
--results-by-company	int	none	Maximum contacts returned per company.
--include-search-contacts / --no-include-search-contacts	bool	unset	Include or exclude contacts from contact search in the results.
--consensus	int	none	Consensus threshold for discovered contacts.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--param	str	none	Extra key=value query parameter forwarded to the SDK (repeatable).
--name	str	none	Filter by contact name (partial match supported).
--summary	str	none	Filter by profile summary text (semantic search).
--negate-summary	str	none	Exclude contacts matching this summary description.
--skills	str	none	Filter by skill (repeatable).
--person-state	str	none	Filter by contact state/region (repeatable).
--persona-id	int	none	Filter by persona ID (repeatable).
--filter-state	str	none	Filter by company state/region (repeatable).
--negate-filter-state	str	none	Exclude contacts at companies in this state (repeatable).
--email-validated	bool	none	Only contacts with (or without) a validated email address.
--has-phone	bool	none	Only contacts with (or without) a phone number.
--has-mobile	bool	none	Only contacts with (or without) a mobile phone number.
--has-linkedin	bool	none	Only contacts with (or without) a LinkedIn profile.
--min-connections	int	none	Minimum LinkedIn connections required.
--inclusion-query-id	str	none	Include only contacts from companies in this saved query (repeatable).
--exclusion-query-id	str	none	Exclude contacts from companies in this saved query (repeatable).
--max-companies	int	none	Maximum number of enriched companies to return; cannot be combined with —max-records.
--params-file	path	none	JSON object of API parameter names (an app form copied over); --param and named flags override its values.

discolike contacts discover --domain stripe.com --seniority vp \
  --results-by-company 5 --include-search-contacts --consensus 2

REST endpoint: Contacts API. SDK: client.contacts.discover().

contacts generate

Generate contacts for target domains from an ICP description (async job). The job is reported with task_family discogen, so poll it with --family discogen.

discolike contacts generate [OPTIONS]

Flag	Type	Default	Description
--icp-text	str	required	ICP description used to generate contacts.
--domain	str	none	Target company domain (repeatable).
--domains-file	path	none	CSV with a domain column, or one domain per line; merged with --domain. One of --domain or --domains-file is required.
--context-mode	str	none	Context mode for generation.
--integration-id	str	none	Integration ID to use for generation.
--search-provider-id	str	none	Search provider ID to use for generation.
--search-context-size	str	none	Search context size for the search provider.
--max-contacts-per-domain	int	none	Maximum contacts generated per domain.
--max-company-records	int	none	Maximum company records to process.
--find-emails	flag	off	Run the email finder over named, email-less rows before the job completes; fills email and email_status. Found addresses bill under the finder’s rules.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--full-domain	str	none	Domain to send as full_domains to the generation job (repeatable).
--partial-domain	str	none	Domain to send as partial_domains to the generation job (repeatable).

discolike contacts generate --icp-text "Heads of RevOps" \
  --domain acme.com --domain beta.com --max-contacts-per-domain 5 --find-emails --wait

Without --find-emails, email is null unless a source showed it verbatim; chain discolike email find-batch on the returned names to look addresses up afterwards.

REST endpoint: Contacts API. SDK: client.contacts.generate().

count

Count companies matching the given filters. Counting is the cheap way to size a query before you run discover.

discolike count [OPTIONS]

The filter set is identical to discover minus --icp-prompt, --domain, --exclude-domain, --exclusion-query-id, --max-records and --offset.

Flag	Type	Default	Description
--phrase-match	str	none	Phrase the company website must contain (repeatable).
--negate-phrase-match	str	none	Negate the --phrase-match filter (repeatable).
--category	str	none	Industry category filter (repeatable).
--negate-category	str	none	Negate the --category filter (repeatable).
--country	str	none	ISO country code filter (repeatable).
--negate-country	str	none	Negate the --country filter (repeatable).
--state	str	none	State or region filter (repeatable).
--negate-state	str	none	Negate the --state filter (repeatable).
--sub-industry	str	none	Sub-industry filter, bare or PARENT/SUB (repeatable).
--negate-sub-industry	str	none	Negate the --sub-industry filter (repeatable).
--geo	str	none	Circular area as lat,lon or lat,lon,radius, e.g. 30.27,-97.74,30mi (repeatable). Radius defaults to 50km.
--bbox	str	none	Bounding box as min_lat,min_lon,max_lat,max_lon (repeatable). Longitudes may wrap the antimeridian.
--lat	float	none	Latitude of the search centre; requires --lon.
--lon	float	none	Longitude of the search centre; requires --lat.
--radius	str	none	Radius around --lat/--lon, e.g. 50km or 30mi. Default 50km.
--employee-range	str	none	Employee count range filter.
--revenue-range	str	none	Revenue range filter.
--business-model	str	none	Business model filter (repeatable).
--negate-business-model	str	none	Negate the --business-model filter (repeatable).
--tech-stack	str	none	Technology stack filter (repeatable).
--negate-tech-stack	str	none	Negate the --tech-stack filter (repeatable).
--min-digital-footprint	int	none	Minimum digital footprint score.
--max-digital-footprint	int	none	Maximum digital footprint score.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--param	str	none	Extra API parameter as KEY=VALUE (comma-separates into a list); see the endpoint reference (repeatable).
--subdomain	str	none	Limit results to this subdomain, up to 20, each at least 3 characters (repeatable).
--negate-subdomain	str	none	Exclude this subdomain, up to 20, each at least 3 characters (repeatable).
--language	str	none	Filter by site language as an ISO 639-1 code, e.g. en, de, pb, zt (repeatable).
--negate-language	str	none	Exclude this site language (repeatable); same values as —language.
--social	str	none	Filter by social platform presence (repeatable): facebook, instagram, linkedin, pinterest, threads, tiktok, twitter, x, yelp, youtube, googleplay, applestore, amazon, vk, bluesky, xing.
--negate-social	str	none	Exclude companies with this social profile (repeatable); same values as —social.
--start-date	str	none	Minimum company start date (YYYY-MM-DD) or range (YYYY-MM-DD,YYYY-MM-DD).
--redirect	bool	none	Include or exclude domains that redirect to another domain.
--exclude-leadgen	bool	none	Exclude suspected lead generation sites (off by default for count).
--params-file	path	none	JSON object of API parameter names (an app form copied over); --param and named flags override its values.

discolike count --category SOFTWARE --country US

{
  "count": 68900
}

discolike count --country DE --employee-range 50-200

{
  "count": 1510010
}

A malformed --param fails before any request is made:

discolike count --param bogus

Usage: discolike count [OPTIONS]
Try 'discolike count --help' for help.
╭─ Error ──────────────────────────────────────────────────────────────────────╮
│ Invalid value: --param must be in KEY=VALUE form, got 'bogus'                │
╰──────────────────────────────────────────────────────────────────────────────╯

REST endpoint: Count API. SDK equivalent: client.count().

discogen

Run DiscoGen research jobs and check status of or cancel any async task (see --family). SDK equivalents live on client.discogen; job handling is covered in Job.

discogen run

Run a DiscoGen research query across company domains (async job).

discolike discogen run [OPTIONS]

Flag	Type	Default	Description
--query	str	required	Research query to run.
--domain	str	none	Company domain to research (repeatable).
--domains-file	path	none	CSV with a domain column, or one domain per line; merged with --domain. One of --domain or --domains-file is required.
--integration-id	str	none	Integration ID to use for the run.
--web-search / --no-web-search	bool	unset	Toggle web search during research.
--context-mode	str	none	Context mode; see the endpoint reference.
--include-x-search / --no-include-x-search	bool	unset	Toggle including X search in the research.
--typed-columns / --no-typed-columns	bool	unset	Let the detector answer yes/no, fixed-set and scale columns with a TypeSafe judgment model.
--include-confidence / --no-include-confidence	bool	unset	Add a confidence column beside each typed column.
--search-provider-id	str	none	Search provider ID to use for web search.
--search-context-size	str	none	Search context size; see the endpoint reference.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike discogen run --query "Recent funding rounds" \
  --domain stripe.com --domain adyen.com --web-search --wait --timeout 600

REST endpoint: DiscoGen API. SDK: client.discogen.process().

discogen run-personas

Run a DiscoGen research query across personas (async job). Identical to run except that --persona-id replaces --domain.

discolike discogen run-personas [OPTIONS]

Flag	Type	Default	Description
--query	str	required	Research query to run.
--persona-id	int	required	Persona ID to research (repeatable).
--integration-id	str	none	Integration ID to use for the run.
--web-search / --no-web-search	bool	unset	Toggle web search during research.
--context-mode	str	none	Context mode; see the endpoint reference.
--include-x-search / --no-include-x-search	bool	unset	Toggle including X search in the research.
--typed-columns / --no-typed-columns	bool	unset	Let the detector answer yes/no, fixed-set and scale columns with a TypeSafe judgment model.
--include-confidence / --no-include-confidence	bool	unset	Add a confidence column beside each typed column.
--search-provider-id	str	none	Search provider ID to use for web search.
--search-context-size	str	none	Search context size; see the endpoint reference.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike discogen run-personas --query "Career history" --persona-id 1 --persona-id 2

REST endpoint: DiscoGen API. SDK: client.discogen.process_personas().

discogen models

List the models available for DiscoGen research.

discolike discogen models [OPTIONS]

Flag	Type	Default	Description
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike discogen models

{
  "models": {
    "openai": [
      {
        "name": "o4-mini",
        "supports_web_search": true
      },
      {
        "name": "gpt-5-pro",
        "supports_web_search": true
      },
      ...
    ],
    ...
  }
}

REST endpoint: DiscoGen API. SDK: client.discogen.models().

discogen status

Show the status and progress of an async task. This is the polling command for every async job in the CLI, not just DiscoGen.

discolike discogen status [OPTIONS] TASK_ID

Flag	Type	Default	Description
TASK_ID (argument)	str	required	Task ID returned when the job was started.
--family	discogen | bulkmatch | contactmatch | segment	discogen	Task family the task_id belongs to (bulk match jobs are ‘bulkmatch’, segment jobs ‘segment’, contact bulk-match ‘contactmatch’).
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike discogen status dg-4

discolike discogen status seg-1 --family segment

Any other --family value is a usage error and exits 2.

SDK: Job.status().

discogen cancel

Cancel an async task. Output is always JSON: {"cancelled": "<task_id>"}.

discolike discogen cancel [OPTIONS] TASK_ID

Flag	Type	Default	Description
TASK_ID (argument)	str	required	Task ID returned when the job was started.
--family	discogen | bulkmatch | contactmatch | segment	discogen	Task family the task_id belongs to (bulk match jobs are ‘bulkmatch’, segment jobs ‘segment’, contact bulk-match ‘contactmatch’).

discolike discogen cancel bm-1 --family bulkmatch

SDK: Job.cancel().

discover

Discover companies matching your ICP and filters.

discolike discover [OPTIONS]

Flag	Type	Default	Description
--icp-prompt	str	none	Natural-language ideal customer profile description.
--domain	str	none	Seed domain for lookalike matching (repeatable).
--phrase-match	str	none	Phrase the company website must contain (repeatable).
--negate-phrase-match	str	none	Negate the --phrase-match filter (repeatable).
--category	str	none	Industry category filter (repeatable).
--negate-category	str	none	Negate the --category filter (repeatable).
--country	str	none	ISO country code filter (repeatable).
--negate-country	str	none	Negate the --country filter (repeatable).
--state	str	none	State or region filter (repeatable).
--negate-state	str	none	Negate the --state filter (repeatable).
--sub-industry	str	none	Sub-industry filter, bare or PARENT/SUB (repeatable).
--negate-sub-industry	str	none	Negate the --sub-industry filter (repeatable).
--geo	str	none	Circular area as lat,lon or lat,lon,radius, e.g. 30.27,-97.74,30mi (repeatable). Radius defaults to 50km.
--bbox	str	none	Bounding box as min_lat,min_lon,max_lat,max_lon (repeatable). Longitudes may wrap the antimeridian.
--lat	float	none	Latitude of the search centre; requires --lon.
--lon	float	none	Longitude of the search centre; requires --lat.
--radius	str	none	Radius around --lat/--lon, e.g. 50km or 30mi. Default 50km.
--employee-range	str	none	Employee count range filter.
--revenue-range	str	none	Revenue range filter.
--business-model	str	none	Business model filter (repeatable).
--negate-business-model	str	none	Negate the --business-model filter (repeatable).
--tech-stack	str	none	Technology stack filter (repeatable).
--negate-tech-stack	str	none	Negate the --tech-stack filter (repeatable).
--min-digital-footprint	int	none	Minimum digital footprint score.
--max-digital-footprint	int	none	Maximum digital footprint score.
--exclude-domain	str	none	Domain to exclude from results (repeatable).
--exclude-domains-file	path	none	CSV with a domain column, or one domain per line; merged with --exclude-domain (100 domains max, checked before the request).
--exclusion-query-id	str	none	Saved query ID whose results are excluded (repeatable).
--max-records	int	none	Maximum number of companies to return.
--offset	int	none	Number of records to skip for pagination.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--param	str	none	Extra API parameter as KEY=VALUE (comma-separates into a list); see the endpoint reference (repeatable).
--subdomain	str	none	Limit results to this subdomain, up to 20, each at least 3 characters (repeatable).
--negate-subdomain	str	none	Exclude this subdomain, up to 20, each at least 3 characters (repeatable).
--language	str	none	Filter by site language as an ISO 639-1 code, e.g. en, de, pb, zt (repeatable).
--negate-language	str	none	Exclude this site language (repeatable); same values as —language.
--social	str	none	Filter by social platform presence (repeatable): facebook, instagram, linkedin, pinterest, threads, tiktok, twitter, x, yelp, youtube, googleplay, applestore, amazon, vk, bluesky, xing.
--negate-social	str	none	Exclude companies with this social profile (repeatable); same values as —social.
--start-date	str	none	Minimum company start date (YYYY-MM-DD) or range (YYYY-MM-DD,YYYY-MM-DD).
--min-similarity	int	none	Minimum similarity score to include (0-99).
--variance	str	none	Result diversity control: LOW, MID_LOW, MEDIUM, MID_HIGH, HIGH, UNRESTRICTED.
--consensus	int	none	Number of top results for the consensus search vector (1-20); higher reduces specificity.
--redirect	bool	none	Include or exclude domains that redirect to another domain.
--exclude-leadgen	bool	none	Exclude suspected lead generation sites (on by default).
--retrieval	bool	none	Enable page data retrieval using the Extract API.
--enhanced	bool	none	Enable AI-powered result enhancement for improved relevance.
--include-search-domains	bool	none	Include the input domains in results (excluded by default).
--auto-icp-text	bool	none	Auto-generate ICP text from the provided domain(s).
--auto-phrase-match	bool	none	Auto-generate phrase matches from ICP text.
--inclusion-query-id	str	none	Saved query ID whose domains are included (repeatable); requires the STARTER plan.
--params-file	path	none	JSON object of API parameter names (an app form copied over); --param and named flags override its values.

Repeatable flags accumulate into a list, so --country US --country CA sends both values.

Every --geo circle, every --bbox box and the --lat/--lon/--radius centre are OR’d together, up to 10 shapes per query: discolike discover --geo 30.27,-97.74,30mi --geo 52.52,13.405 searches both areas.

--param is the escape hatch for filters that have no dedicated flag. It splits on the first =, and a value containing a comma becomes a list - --param social=linkedin,youtube sends social=linkedin&social=youtube. A value with no = is a usage error (exit 2); a value the API schema rejects exits 2 before the request is sent. Named flags win over --param for the same key: --country DE --param country=US sends country=DE.

--params-file form.json loads a whole request from a JSON object keyed by API parameter names, the shape you get by copying an app form over; it sits under --param, which sits under the named flags. The file is validated against the request schema before anything is sent, so a bad enum value exits 2 with no request made.

discolike discover --params-file form.json --param consensus=3 --variance MID_HIGH --max-records 100

discolike discover --icp-prompt "B2B payments infrastructure" \
  --country US --employee-range 50-200 --max-records 3

discolike discover --domain stripe.com --domain adyen.com \
  --negate-country RU --exclude-domain stripe.com --max-records 3

REST endpoint: Discover API. SDK equivalent: client.discover().

email

Find work email addresses: submit single or batch find jobs, poll them, and fetch results. Only proven addresses bill; catch-all and pattern guesses are free.

Every submission is asynchronous. find and find-batch return an ID immediately; add --wait to block until the result is ready, or poll later with email job and email results. SDK equivalents live on client.email.

Email jobs use their own IDs and their own polling commands, separate from discogen status.

email find

Submit a single email find job (async); only a proven address bills.

discolike email find [OPTIONS] FIRST_NAME LAST_NAME DOMAIN

Flag	Type	Default	Description
FIRST_NAME (argument)	str	required	First name of the person.
LAST_NAME (argument)	str	required	Last name of the person.
DOMAIN (argument)	str	required	Company domain to search, e.g. acme.com.
--known-pattern	str	none	Known email local-part pattern for this domain, e.g. first.last.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

Without --wait the command prints the job ID and the command that polls it. The payload is a single object, so it always prints as JSON regardless of --format.

discolike email find Patrick Collison stripe.com

{
  "job_id": "98e10684-a270-4167-85a6-dc638bcf3791",
  "hint": "poll with: discolike email job 98e10684-a270-4167-85a6-dc638bcf3791"
}

With --wait the command blocks and prints the enumeration result instead, writing status: <value> lines to stderr on each poll.

discolike email find Patrick Collison stripe.com --wait --timeout 120

Pass --known-pattern when you already know how a domain builds its addresses; it is omitted from the request entirely when you leave it off.

discolike email find Ada Lovelace acme.com --known-pattern first.last --wait

REST endpoint: Email Find API. SDK: client.email.find().

email find-batch

Submit an email find batch from a CSV file and/or inline contacts (async).

discolike email find-batch [OPTIONS]

Flag	Type	Default	Description
--contacts-file	path	none	Path to a CSV file with first_name,last_name,domain columns (max 500 contacts per batch).
--contact	str	none	Inline contact as “first_name,last_name,domain” (repeatable).
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

Provide --contacts-file, one or more --contact, or both; the two sources are concatenated, file rows first. Supplying neither exits 2, and so does a total above 500 contacts or a CSV missing any of the three required columns.

discolike email find-batch --contacts-file people.csv \
  --contact "Ada,Lovelace,acme.com" --wait --timeout 600

Without --wait you get the batch ID and the command that fetches it:

{
  "batch_id": "7f6f0a4e-52f7-4f3e-9c2d-0b8d3f1c9a11",
  "hint": "fetch with: discolike email results 7f6f0a4e-52f7-4f3e-9c2d-0b8d3f1c9a11"
}

With --wait the progress lines on stderr read progress: <completed>/<total> completed, <failed> failed.

REST endpoint: Email Find API. SDK: client.email.find_batch().

email results

Fetch results for an email find or verify batch.

discolike email results [OPTIONS] BATCH_ID

Flag	Type	Default	Description
BATCH_ID (argument)	str	required	Batch ID returned by discolike email find-batch.
--kind	find | verify	find	Batch kind: find or verify (verify batches are created by the DiscoLike app).
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

Without --wait the command fetches one snapshot and prints it even if the batch is still running, so completed and total can disagree. Any other --kind value is a usage error and exits 2.

--kind selects how each result is decoded: find reads them as enumeration results, verify as validation results. Pass the --kind that matches the batch.

discolike email results 7f6f0a4e-52f7-4f3e-9c2d-0b8d3f1c9a11 --wait

discolike email results <batch_id> --kind verify

REST endpoint: Email Find API. SDK: client.email.batch().

email job

Poll a single email find job: print its current status, or block with --wait. The command is find-only and has no --kind flag; rehydrating a verify job needs the SDK’s client.email.job(job_id, kind="verify").

discolike email job [OPTIONS] JOB_ID

Flag	Type	Default	Description
JOB_ID (argument)	str	required	Job ID returned by discolike email find.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

Without --wait you get the job wrapper: lifecycle status on the outside, the enumeration result nested under result:

discolike email job 98e10684-a270-4167-85a6-dc638bcf3791

{
  "job_id": "98e10684-a270-4167-85a6-dc638bcf3791",
  "status": "completed",
  "result": {
    "first_name": "patrick",
    "last_name": "collison",
    "domain": "stripe.com",
    "status": "catch_all_pattern",
    "result": {
      "email": "patrick.collison@stripe.com",
      "pattern": "first.last",
      "tier": 1,
      "smtp_code": 0,
      "valid": false
    },
    "is_catch_all": true,
    "mx_host": "aspmx3.googlemail.com",
    "provider": "Everything Else",
    "attempts": 0,
    "duration_ms": 1473,
    "error": null
  },
  "error": null
}

With --wait the wrapper is stripped and only the enumeration result is printed:

discolike email job 98e10684-a270-4167-85a6-dc638bcf3791 --wait --timeout 120

{
  "first_name": "patrick",
  "last_name": "collison",
  "domain": "stripe.com",
  "status": "catch_all_pattern",
  "result": {
    "email": "patrick.collison@stripe.com",
    "pattern": "first.last",
    "tier": 1,
    "smtp_code": 0,
    "valid": false
  },
  "is_catch_all": true,
  "mx_host": "aspmx3.googlemail.com",
  "provider": "Everything Else",
  "attempts": 0,
  "duration_ms": 1473,
  "error": null
}

Polling is free; only the original find can bill, and only when it proves an address. The two status fields mean different things: the outer one is the job lifecycle, the inner one is the enumeration outcome. See Email Find for every value the inner status can take.

REST endpoint: Email Find API. SDK: client.email.job().

extract

Extract page content from a URL.

discolike extract [OPTIONS] URL

Flag	Type	Default	Description
URL (argument)	str	required	Page URL to extract content from.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--domain	str	none	Bare domain to extract instead of URL (alias for https://DOMAIN, served from cache).

discolike extract https://stripe.com/about

{
  "text": "Stripe | Financial Infrastructure to Grow Your Revenue Products Solutions Developers Resources Pricing Sign in Start now Contact sales ...",
  "language": "en"
}

REST endpoint: Extract API. SDK equivalent: client.companies.extract().

llm-providers

Manage BYOK LLM provider integrations (OpenAI, Anthropic, custom endpoints). SDK equivalents live on client.llm_providers; the REST surface is the LLM Providers API.

llm-providers list

List LLM provider integrations for the organization.

discolike llm-providers list [OPTIONS]

Flag	Type	Default	Description
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike llm-providers list

{
  "providers": [
    {
      "integration_id": "1f5e8466-fda4-4d4e-88eb-4b0886c37003",
      "integration_name": "grok-4.5-latest",
      "provider": "xai",
      "api_key": "*****",
      "model_name": "xai/grok-4.5-latest",
      "base_url": null,
      "supports_web_search": true,
      "input_cost_per_token": 2e-06,
      "output_cost_per_token": 6e-06,
      "is_default": true,
      "model_deprecated": false
    },
    ...
  ]
}

llm-providers create

Create an LLM provider integration. No --format flag.

discolike llm-providers create [OPTIONS]

Flag	Type	Default	Description
--name	str	required	User-friendly name for the integration.
--provider	str	required	LLM provider name: openai, anthropic, or custom. Free-form text - not validated against a list.
--api-key	str	required	API key for the provider.
--model-name	str	required	Default model name for this integration.
--base-url	str	none	Endpoint URL (required for the custom provider).

discolike llm-providers create --name OpenAI --provider openai \
  --api-key sk-... --model-name gpt-4o

llm-providers get

Get a single LLM provider integration by ID.

discolike llm-providers get [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the LLM provider integration to fetch.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike llm-providers get 1f5e8466-fda4-4d4e-88eb-4b0886c37003

{
  "integration_id": "1f5e8466-fda4-4d4e-88eb-4b0886c37003",
  "integration_name": "grok-4.5-latest",
  "provider": "xai",
  "api_key": "*****",
  "model_name": "xai/grok-4.5-latest",
  "base_url": null,
  "supports_web_search": true,
  "input_cost_per_token": 2e-06,
  "output_cost_per_token": 6e-06,
  "is_default": true,
  "model_deprecated": false
}

llm-providers update

Update an LLM provider integration. --name, --provider and --model-name are required on every update. Omitting --api-key keeps the stored key. No --format flag.

discolike llm-providers update [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the LLM provider integration to update.
--name	str	required	User-friendly name for the integration.
--provider	str	required	LLM provider name: openai, anthropic, or custom.
--model-name	str	required	Default model name for this integration.
--api-key	str	none	New API key (omit to keep the stored key).
--base-url	str	none	Endpoint URL (required for the custom provider).

discolike llm-providers update 1f5e8466-fda4-4d4e-88eb-4b0886c37003 \
  --name Anthropic --provider anthropic --model-name claude-sonnet-4-5

llm-providers delete

Delete an LLM provider integration (admin only). Prints {"deleted": "<integration_id>"}.

discolike llm-providers delete [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the LLM provider integration to delete.

discolike llm-providers delete 1f5e8466-fda4-4d4e-88eb-4b0886c37003

llm-providers set-default

Set an LLM provider integration as the organization default (admin only). The default is used whenever a command omits --integration-id.

discolike llm-providers set-default [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the integration to make the organization default.

discolike llm-providers set-default 1f5e8466-fda4-4d4e-88eb-4b0886c37003

llm-providers test-connection

Test a provider configuration before saving it. Nothing is persisted. No --format flag.

discolike llm-providers test-connection [OPTIONS]

Flag	Type	Default	Description
--name	str	cli-test	Name to label the test configuration.
--provider	str	required	LLM provider name: openai, anthropic, or custom.
--api-key	str	required	API key for the provider.
--model-name	str	required	Model name to test against.
--base-url	str	none	Endpoint URL (required for the custom provider).

discolike llm-providers test-connection --provider openai \
  --api-key sk-... --model-name gpt-4o

match

Match a company name to a domain, or bulk-match a CSV of names.

discolike match [OPTIONS] [NAME]

Flag	Type	Default	Description
NAME (argument)	str	none	Company name to match to a domain. Optional, but exactly one of NAME or --file is required.
--phone	str	none	Phone number to disambiguate the match.
--city	str	none	City to disambiguate the match.
--state	str	none	State or region to disambiguate the match.
--country	str	none	Country to disambiguate the match.
--zip-code	str	none	ZIP or postal code to disambiguate the match.
--strict / --no-strict	bool	unset	Toggle strict matching. Omitted from the request entirely when neither form is passed.
--local-mode / --no-local-mode	bool	unset	Toggle local matching mode. Omitted from the request entirely when neither form is passed.
--file	path	none	CSV of company names to bulk-match as an async job (instead of NAME).
--name-column	str	name	Column in --file that holds company names.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--min-match-confidence	int	none	Minimum match_confidence (inclusive, 50-100) a match must have to be returned.
--phone-column	str	none	Column in —file that holds phone numbers.
--city-column	str	none	Column in —file that holds cities.
--state-column	str	none	Column in —file that holds states.
--country-column	str	none	Column in —file that holds country codes.
--zip-code-column	str	none	Column in —file that holds zip codes.

Passing both NAME and --file, or neither, is a usage error (exit 2). With NAME the call is synchronous; with --file it starts a bulkmatch job.

discolike match "Stripe Inc"

{
  "query": {
    "name": "Stripe Inc",
    "country": null,
    "state": null,
    "city": null,
    "zip": null,
    "phones": null
  },
  "matches": [
    {
      "domain": "stripe.com",
      "name": "Stripe",
      "status": {
        "status": "active",
        "confidence": 0.89
      },
      "score": 701,
      "start_date": "2011-01-04",
      "end_date": null,
      "address": {
        "street": "354 Oyster Point Blvd",
        "city": "South San Francisco",
        "state": "California",
        "zip": "94080",
        "country": "US"
      },
      "phones": [
        "+16504279276",
        "+18889262289"
      ],
      ...
    }
  ]
}

Disambiguate with location fields when the name is ambiguous:

discolike match "Acme" --city Boston --state MA --country US --strict

Bulk-match a CSV and wait for the job to finish:

discolike match --file names.csv --name-column company --wait --timeout 600

REST endpoints: Match API for a single name, Bulk Match API for --file. SDK equivalents: client.match.company() and client.match.bulk().

queries

Manage saved queries and exclusion lists for reusable targeting. SDK equivalents live on client.queries; the REST surface is the Queries API.

queries list

List saved queries.

discolike queries list [OPTIONS]

Flag	Type	Default	Description
--max-records	int	none	Maximum number of saved queries to return.
--offset	int	none	Number of records to skip for pagination.
--action	str	none	Filter by query action, e.g. discover. Free-form text - not validated against a list.
--tag	str	none	Filter by tag (repeatable).
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike queries list --max-records 1 --action discover

{
  "results": [
    {
      "query_id": "e15abe19-a8dc-41ca-8c79-38047ede6450",
      "query_name": "USA law firms in United States",
      "action": "discover",
      "user_name": "George Rekouts",
      "mtime": "2026-08-19T20:25:35.912744",
      "domains": [
        "legalpages.co.uk",
        "myfamilyattorneys.com",
        ...
      ],
      ...
    }
  ]
}

queries create-exclusion-list

Create a named exclusion list of domains and/or persona IDs. No --format flag.

discolike queries create-exclusion-list [OPTIONS]

Flag	Type	Default	Description
--name	str	required	Name for the new exclusion list.
--domain	str	none	Domain to exclude (repeatable).
--domains-file	path	none	CSV with a domain column, or one domain per line; merged with --domain.
--persona-id	int	none	Persona ID to exclude (repeatable).
--tag	str	none	Tag to attach to the list (repeatable).

discolike queries create-exclusion-list --name "Current customers" \
  --domain acme.com --domain beta.com --tag crm

A page of 10,000 discover results goes in from its CSV:

discolike queries create-exclusion-list --name "tam-round-1" --domains-file round1.csv --tag tam

Feed the resulting query ID back into discovery with discolike discover --exclusion-query-id <id>.

queries save-results

Save result rows from a file as a reusable saved query. No --format flag.

discolike queries save-results [OPTIONS]

Flag	Type	Default	Description
--input	path	required	Path to a .json (list of row objects) or .csv (header row) file.
--name	str	required	Name for the saved query.
--action	discover | segment | contacts | append | match	required	Underlying action: discover, segment, contacts, append, or match.
--domain-column	str	domain	Column holding domains.
--tag	str	none	Tag to attach (repeatable).

The file suffix decides the parser: .csv (case-insensitive) is read with a CSV reader and every value stays a string; anything else is parsed as JSON. A missing file or invalid JSON is a usage error (exit 2), as is an --action outside the five choices.

discolike queries save-results --input rows.json --name "Q3 targets" --action discover

discolike queries save-results --input rows.csv --name "Q3 targets" \
  --action discover --domain-column website --tag q3

queries update

Rename a saved query and/or update its tags. No --format flag.

discolike queries update [OPTIONS] QUERY_ID

Flag	Type	Default	Description
QUERY_ID (argument)	str	required	ID of the saved query to update.
--name	str	none	New name for the saved query.
--tag	str	none	Tag to set on the query (repeatable).

discolike queries update e15abe19-a8dc-41ca-8c79-38047ede6450 --name "Renamed" --tag hot

queries delete

Delete a saved query. No options beyond the argument; prints {"deleted": "<query_id>"}.

discolike queries delete [OPTIONS] QUERY_ID

Flag	Type	Default	Description
QUERY_ID (argument)	str	required	ID of the saved query to delete.

discolike queries delete e15abe19-a8dc-41ca-8c79-38047ede6450

search-providers

Manage BYOK web search provider integrations (Tavily, Serper, etc.). SDK equivalents live on client.search_providers; the REST surface is the Search Providers API.

search-providers list

List search provider integrations for the organization.

discolike search-providers list [OPTIONS]

Flag	Type	Default	Description
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike search-providers list

{
  "providers": [
    {
      "integration_id": "fd467f7f-11c1-4475-923a-f2c3341e41bb",
      "integration_name": "Serper",
      "provider": "serper",
      "search_model": "serper/search",
      "encrypted_api_key": null,
      "api_key": "*****",
      "base_url": null,
      "is_default": true,
      "cost_per_query": 0.001
    },
    ...
  ]
}

search-providers create

Create a search provider integration (connectivity is validated first). No --format flag.

discolike search-providers create [OPTIONS]

Flag	Type	Default	Description
--name	str	required	User-friendly name for the integration.
--provider	str	required	Search provider key, e.g. tavily or serper.
--search-model	str	required	Search model key, e.g. tavily/search.
--api-key	str	none	Provider API key (omit for free providers).
--base-url	str	none	Custom endpoint URL for a LiteLLM proxy.

discolike search-providers create --name Tavily --provider tavily \
  --search-model tavily/search --api-key tvly-...

search-providers update

Update a search provider integration. --name, --provider and --search-model are required on every update - this is a full replace, not a patch. No --format flag.

discolike search-providers update [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the search provider integration to update.
--name	str	required	User-friendly name for the integration.
--provider	str	required	Search provider key, e.g. tavily or serper.
--search-model	str	required	Search model key, e.g. tavily/search.
--api-key	str	none	New provider API key (omit to keep the stored key).
--base-url	str	none	Custom endpoint URL for a LiteLLM proxy.

discolike search-providers update fd467f7f-11c1-4475-923a-f2c3341e41bb \
  --name Serper --provider serper --search-model serper/search

search-providers delete

Delete a search provider integration (admin only). Prints {"deleted": "<integration_id>"}.

discolike search-providers delete [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the search provider integration to delete.

discolike search-providers delete fd467f7f-11c1-4475-923a-f2c3341e41bb

search-providers set-default

Set a search provider integration as the organization default (admin only).

discolike search-providers set-default [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the integration to make the organization default.

discolike search-providers set-default fd467f7f-11c1-4475-923a-f2c3341e41bb

search-providers clear-default

Remove the default flag from a search provider integration (admin only).

discolike search-providers clear-default [OPTIONS] INTEGRATION_ID

Flag	Type	Default	Description
INTEGRATION_ID (argument)	str	required	ID of the integration currently set as default.

discolike search-providers clear-default fd467f7f-11c1-4475-923a-f2c3341e41bb

search-providers models

List available search models grouped by provider, with cost per query. Use this to find the exact --search-model value for create and update.

discolike search-providers models [OPTIONS]

Flag	Type	Default	Description
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike search-providers models

{
  "models": {
    "apiserpent": [
      {
        "name": "apiserpent/search/google",
        "cost_per_query": 0.0006
      },
      {
        "name": "apiserpent/search/bing",
        "cost_per_query": 0.0006
      },
      ...
    ],
    ...
  }
}

segment

Auto-segment a list of domains (async job). The task family is segment.

discolike segment [OPTIONS]

Flag	Type	Default	Description
--domain	str	none	Domain to segment (repeatable).
--file	path	none	CSV of domains to segment (instead of --domain).
--domain-column	str	none	Column in --file that holds domains.
--max-segments	int	none	Maximum number of segments to produce. 2-20 with --domain, 2-100 with --file.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).
--query-id	str	none	Saved query ID whose domains are included alongside the file/—domain ones (repeatable).

Pass exactly one of --domain or --file; both or neither is a usage error (exit 2).

discolike segment --domain acme.com --domain beta.com --max-segments 3

Without --wait the command prints a job handle - task_id, task_family (segment), and a hint string with the exact discolike discogen status command to poll it - and exits immediately.

discolike segment --file domains.csv --domain-column domain --wait --timeout 600

REST endpoint: Segment API. SDK equivalent: client.segment() for --domain, client.segment_file() for --file.

signup

Create a DiscoLike account for a person. No login needed; they confirm by email. This is the one command that works before discolike auth login.

discolike signup [OPTIONS]

Flag	Type	Default	Description
--email	str	required	The person’s work email. Becomes the login.
--first-name	str	required	1-40 characters. Must contain a letter and no angle brackets or control characters.
--last-name	str	required	Same rules as --first-name.
--agent	str	none	Agent or framework name to record with the signup. Defaults to discolike-cli/<version>.
--yes, -y	flag	off	Skip the confirmation when signing up a different email than last time.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

discolike signup --email jane@acme.com --first-name Jane --last-name Doe

{
  "status": "created",
  "email": "jane@acme.com",
  "org_domain": "acme.com",
  "org_status": "created",
  "next_step": "A confirmation email was sent to jane@acme.com. Log in at https://app.discolike.com. Google or Microsoft sign-in with this email also works, no password needed."
}

Relay next_step to the person: they confirm by email, log in at app.discolike.com, pick a plan, and issue an API key for discolike auth login.

The command remembers the email it signed up. Running it again with a different email asks for confirmation first, and on a non-TTY exits 1 telling you to re-run with --yes, so an agent cannot quietly create a second account.

REST endpoint: Signup API. SDK equivalent: signup(). Per-agent instructions: Agent signup guide.

validate-icp

Validate a list of domains against an ICP definition (async job).

discolike validate-icp [OPTIONS]

Flag	Type	Default	Description
--icp	str	required	ICP definition text to validate the domains against.
--domain	str	none	Domain to validate (repeatable).
--domains-file	path	none	CSV with a domain column, or one domain per line (instead of --domain).
--file	path	none	Older spelling of --domains-file.
--context-mode	str	none	Context mode; see the endpoint reference.
--integration-id	str	none	Integration ID to use for the validation.
--web-search / --no-web-search	bool	unset	Toggle web search during validation.
--search-provider-id	str	none	Search provider ID to use for web search.
--wait	flag	off	Block until the job finishes, streaming progress to stderr.
--timeout	float	900.0	Max seconds to wait with --wait.
--format	str	none	Output format: json or table (table auto-selected on a TTY; falls back to JSON for non-tabular data).

Pass exactly one of --domain or --domains-file (--file is the older spelling of the same flag); both or neither is a usage error (exit 2). The file is a CSV with a domain column, or plain text with one domain per line; domains are lower-cased, stripped of www., and de-duplicated.

discolike validate-icp --icp "Mid-market SaaS selling to finance teams" \
  --domain acme.com --domain beta.com

discolike validate-icp --icp "Mid-market SaaS" --domains-file domains.txt \
  --web-search --wait --timeout 600

REST endpoint: Validate API. SDK equivalent: client.validate_icp().