Skip to main content
Glama

Create a CSV export

argorant_create_export
Destructive

Create a CSV file of the people who match a search, with names, work emails and profile links. Use it when the user wants a spreadsheet of more people than a reveal shows. For a saved list use argorant_export_list, and for a few people in the chat use argorant_reveal_people. Each person with a working email costs 1 credit, and people exported before are skipped unless exclude_previously_exported is false. Phone numbers are only included when include_phones is true. They cost 10 credits per phone number returned, nothing if none is found. Ask the user before spending. Phone numbers need the Pro plan or higher. Tell the user what it can cost and get their approval before you start, because asking again can start a second export. The file is made in the background; follow it with argorant_export_status, or argorant_export_batch_status when a large search is split into several files. A search with exclusions, title or keyword match choices, or the gender, income, range or place filters exports up to 50,000 people at a time; for more, save it with argorant_create_list and export that list. It takes the same filters and defaults as argorant_count_people. More in the docs at https://docs.argorant.com/docs/agents/tools#reveal-enrich-export

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "argorant_create_export"
  }
}

Input Schema

NameRequiredDescriptionDefault
qNoFree text for the search, such as "dental clinics" or a person's job and place. A request such as "software companies selling to restaurants" is read as company_category and sells_to. Prefer the dedicated filters when you know them.
cityNoCities by name. A comma means any of them.
limitNoMost people to put in the file (default 1,000).
stateNoStates or provinces by name or code, such as Texas or TX. A comma means any of them.
titleNoJob titles to find. A comma means any of them. Short forms work both ways, so CFO also finds Chief Financial Officer.
genderNoGender, female or male, based on first name and country. Names used for both women and men are left out. Not available for lending audiences.
radiusNoDistance around the place, between 0.5 and 500.
countryNoCountries, or regions such as Europe, EMEA, DACH, Nordics, Benelux, APAC, LATAM, North America or GCC. A comma means any of them. Regions can overlap. Egypt is in both Middle East and Africa, Mexico is in both North America and LATAM, and Turkey counts as Middle East, not Europe. A person is counted once, even when several of the regions you name include their country.
industryNoIndustries by name, such as Software or Hospitals and Health Care. A comma means any of them.
keywordsNoWords that describe what the company does, such as fintech, MSP or freight forwarding. A comma means any of them. They match the company's name, tags, industry and description, not people's own skills, unless you set where keywords look to people or any.
near_latNoLatitude of the place to search around. Use it with near_lon and radius.
near_lonNoLongitude of the place to search around. Use it with near_lat and radius.
sells_toNoWho the company sells to, used with company_category. One of restaurants, construction, dentists, salons, auto_repair, hotels, logistics, trades, retailers or smb. It finds the companies that sell, never their customers.
geographyNoThe same as country. Countries or regions such as Europe, DACH, Nordics or North America.
has_emailNotrue (the default) looks only at people with a work email, like every count, preview, export and list. false also includes people without one.
has_phoneNotrue keeps only people with a phone number on file. Phone numbers themselves are only shown when paid for.
seniorityNoHow senior the person is. Pick one or more.
near_placeNoA place to search around, written out as a city or a postal code, for example Denver, Colorado or 10115 Berlin. Use it with radius.
departmentsNoThe person's department. Pick one or more.
founded_maxNoCompanies founded in this year or earlier, such as 2020.
founded_minNoCompanies founded in this year or later, such as 2015. Companies whose founding year is not known are left out.
radius_unitNoUnit of the distance, miles (mi, the default) or kilometers (km).
revenue_maxNoLargest yearly revenue in US dollars, such as 50M.
revenue_minNoSmallest yearly revenue in US dollars, such as 1M. A company counts when its revenue range overlaps yours.
title_matchNoHow a job title must match. words (the default) needs every word as a whole word, so Data does not match Database. exact needs the whole title to be one of the values, so CEO does not match Founder & CEO. contains finds the text anywhere, also inside a word, and is slower.
company_nameNoCompany names. A comma means any of them. Prefer company_domain when you know the website.
exclude_cityNoCities to leave out. A comma means any of them.
has_linkedinNotrue keeps only people with a profile link.
revenue_bandNoYearly company revenue in US dollars, as bands. Pick one or more.
technologiesNoTechnologies the company uses, such as Shopify, HubSpot or AWS. A comma means any of them.
country_matchNoWhose place a country means. person (the default) counts people by where they are based and leaves out people whose own place is not known. person_or_company also counts them by their company's country.
employees_maxNoLargest number of employees, such as 500.
employees_minNoSmallest number of employees, such as 50. A company counts when its size range overlaps yours.
exclude_stateNoStates or provinces to leave out. A comma means any of them.
exclude_titleNoJob titles to leave out, matched the same way as title. A comma means any of them.
keyword_scopeNoWhere keywords look. company (the default) looks only at what the company does: its name, tags, industry and description, so a search for MSP finds people at managed service providers. Use people only when the user wants people who have a skill themselves, such as people who know Python or Salesforce admins at any company; it looks only at the person's own skills and profile. any looks at both, plus job titles. Without a value, company is used.
radius_targetNoWhose location counts. any (the default) is where the person is based, or where their company is when their own place is not known. person is only where the person is based, company only where their company is based.
verified_onlyNoDeprecated. It has no effect on the search and will be removed. Leave it out.
company_domainNoCompany website domains, such as stripe.com. A comma means any of them.
employee_rangeNoNumber of employees at the company: size bands (pick one or more), or one text value with a range like "1-500".
include_phonesNotrue also looks up phone numbers, at the phone price in the description. Off by default.
industry_matchNoHow industries match. any (the default) finds the industry anywhere in what the company is tagged with. primary only counts the company's main industry, so Software and Internet leaves out a retailer or a bank that also carries a software tag.
company_countryNoCountries or regions where the company has its head office, with the same names as country. Use it with country to get people in a country at companies based there. For people it works together with a list of company websites (company_domain) for now.
email_when_doneNotrue emails the user when the file is ready.
exclude_countryNoCountries or regions to leave out, matched the same way as country. A comma means any of them.
company_categoryNoWhat kind of company sells, used with sells_to. One of software, payments, pos_payments, business_fintech, accounting_software, payroll_hr_software, formation, association, marketplace or investors.
exclude_industryNoIndustries to leave out, matched the same way as industry. A comma means any of them.
exclude_keywordsNoWords to leave out, matched the same way as keywords. A comma means any of them.
exclude_seniorityNoSeniority levels to leave out. Pick one or more.
exclude_assistantsNoOn by default. A job title search leaves out assistant titles such as assistant to, executive assistant or personal assistant. Set it to false to keep them. The answer lists what it left out under applied_defaults, and a saved list keeps the setting.
business_email_onlyNoOnly people whose email is at their company's own domain. Off unless you set it, so private addresses are left out only on request. When it is on, the answer says so under applied_filters.
exclude_departmentsNoDepartments to leave out. Pick one or more.
exclude_title_matchNoHow the titles to leave out must match, with the same choices. When it is not set, the title choice is used.
exclude_company_domainNoCompany website domains to leave out. A comma means any of them.
exclude_previously_exportedNotrue (the default) skips people this account exported before. false puts them in the file again.

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.6/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Beyond the annotations, the description discloses critical behavioral traits: each person with a working email costs 1 credit, phone numbers cost 10 credits and require Pro plan or higher, previously exported people are skipped by default, and the file is built in the background. It also warns that asking again can start a second export, which is a non-obvious side effect not conveyed by the annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is front-loaded with the core purpose and alternatives, and most sentences carry operational value. However, the cost-approval warning is repeated ('Ask the user before spending' and 'Tell the user what it can cost and get their approval before you start'), which is slightly redundant in an otherwise efficient definition.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 55-parameter, credit-spending export tool with an output schema, the description covers cost, plan requirements, defaults, limit behavior, follow-up status tools, and alternatives. It gives the agent all the information needed to invoke the tool correctly and follow up on the background job.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so all 55 parameters already have detailed descriptions. The tool description only adds a pointer that it 'takes the same filters and defaults as argorant_count_people,' which is useful context but does not add meaning beyond what the schema provides for each parameter. Baseline 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The first sentence states a specific verb and resource: 'Create a CSV file of the people who match a search, with names, work emails and profile links.' It explicitly distinguishes this tool from argorant_export_list (for saved lists) and argorant_reveal_people (for a few people), so an agent can select it without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives explicit when-to-use guidance: 'Use it when the user wants a spreadsheet of more people than a reveal shows,' and names the exact alternatives for other scenarios. It also states the 50,000-person limit with certain filters and directs the agent to argorant_create_list for larger exports, covering the main decision points.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.