generated: '2026-08-11'
method: derived
source: openapi/_original/*.json (27 operations across 26 specs) + live probes of mcp.underscoredone.com
status: candidate
summary: _done runs NO MCP server. mcp.underscoredone.com answers 200 only because *.underscoredone.com
  is a wildcard that serves the marketing homepage for every label (verified against zzz-not-real-abc123.underscoredone.com,
  byte-identical), and a JSON-RPC tools/list POST to it returns 405 Method Not Allowed. The tool list
  below is DERIVED by API Evangelist from the 26 published OpenAPI specs as a candidate forge target.
  No MCPServer pointer is wired into apis.yml, because no MCP server exists.
hosted_server: null
probes:
- url: https://mcp.underscoredone.com/
  method: GET
  status: 200
  verdict: FALSE POSITIVE — wildcard DNS, marketing homepage HTML, not an MCP endpoint.
- url: https://mcp.underscoredone.com/
  method: POST tools/list
  status: 405
  verdict: Method Not Allowed.
- url: https://mcp.underscoredone.com/mcp
  method: POST tools/list
  status: 405
  verdict: Method Not Allowed.
- url: https://underscoredone.com/mcp
  method: GET
  status: 404
- url: https://underscoredone.com/sse
  method: GET
  status: 404
forge_notes:
- Every tool is a single POST with a JSON body — a 1:1 OpenAPI-operation-to-tool mapping is exact, no
  composition needed.
- Tool names below are namespaced by SERVICE because operationIds collide across specs (handler_lookup_post
  appears three times, handler_check_post four).
- 'Any forged server must implement the x402 payment leg: the first call returns 402 with a payment-required
  challenge and no result.'
- directory_submission_lite is the only side-effecting tool (POST /submit places a paid order); it should
  be gated behind confirmation in any agent runtime.
tool_count: 27
tools:
- name: asn_lookup
  title: ASN Lookup
  description: Finds the network owner and details behind an IP address or ASN
  rest:
    method: POST
    server: https://asn-lookup.underscoredone.com
    path: /lookup
    operationId: handler_lookup_post
  input_schema_source: openapi/_original/done-asn-lookup-openapi.json#/components/schemas/Request
  inputs:
  - name: query
    type: string
    required: true
    description: An internet address such as 8.8.8.8 or 2001:4860:4860::8888, or a network reference such
      as AS15169 or 15169, that you want information about.
  price_usd: 0.01
  ai_instructions: Send a query that is either a plain IP address (like 8.8.8.8) or an ASN reference (like
    AS15169 or 15169). The tool figures out which one it is automatically. It looks up live data from
    the RIPEstat public information service (no key required) and returns the owning network's details.
    If the query is invalid or nothing is found, it returns a clear error message instead of guessing.
- name: bimi_checker
  title: BIMI Checker
  description: Checks if a domain's email logo branding setup is valid and ready to display.
  rest:
    method: POST
    server: https://bimi-checker.underscoredone.com
    path: /check
    operationId: handler_check_post
  input_schema_source: openapi/_original/done-bimi-checker-openapi.json#/components/schemas/Request
  inputs:
  - name: domains
    type: array
    required: true
    description: List of website addresses to check, such as example.com. Up to 10 per request. Do not
      include http:// or https://.
  - name: selector
    type: string
    required: false
    description: The label used to find the branding settings, almost always 'default'. Only change this
      if the domain uses a custom label.
  - name: check_vmc
    type: boolean
    required: false
    description: Whether to also verify the logo's security certificate. Turn off for a quicker check.
  price_usd: 0.01
  ai_instructions: 'Call this when a user asks whether a domain''s email logo (BIMI) setup will work,
    why a brand logo is not showing in inboxes, or to audit email branding for one or more domains. Submit
    domains without http/https prefixes. Limit to 10 domains per call. Set check_vmc to false for a faster,
    cheaper check when certificate validation is not needed. Read the issues list per domain to explain
    failures in '
- name: cors_header_checker
  title: CORS Header Checker
  description: Check which cross-origin permissions a URL grants for a given origin
  rest:
    method: POST
    server: https://cors-header-checker.underscoredone.com
    path: /check
    operationId: handler_check_post
  input_schema_source: openapi/_original/done-cors-header-checker-openapi.json#/components/schemas/CheckRequest
  inputs:
  - name: url
    type: string
    required: true
    description: The full web address of the endpoint you want to check — must start with http:// or https://.
  - name: origin
    type: any
    required: false
    description: The website address you want to test cross-origin access from, for example https://myapp.com.
      Defaults to https://example.com if you leave it out.
  price_usd: 0.01
  ai_instructions: POST a JSON body with a required 'url' field (the endpoint to probe) and an optional
    'origin' field (defaults to https://example.com). The response always includes 'cors_enabled' (boolean,
    true if any CORS header was found), 'origin_allowed' (boolean, true if the tested origin would be
    permitted by the browser), and a 'headers' object with all six Access-Control-* values as strings
    or null when ab
- name: cpi_report_us
  title: US CPI Data & Inflation Report
  description: Latest, single-month, or ranged US CPI inflation data from BLS with plain-English summary.
  rest:
    method: POST
    server: https://cpi-report-us.underscoredone.com
    path: /cpi
    operationId: handler_cpi_post
  input_schema_source: openapi/_original/done-cpi-report-us-openapi.json#/components/schemas/Request
  inputs:
  - name: period
    type: any
    required: false
    description: A single month to look up, written as year-month, like 2024-03. Leave this out if you
      want the newest data or a range instead.
  - name: start
    type: any
    required: false
    description: The first month of a historical range, written as year-month, like 2023-01. Leave out
      to start from the earliest available month.
  - name: end
    type: any
    required: false
    description: The last month of a historical range, written as year-month, like 2024-12. Leave out
      to end at the newest available month.
  price_usd: 0.01
  ai_instructions: Send an empty body for the newest available inflation report. Send 'period' as YYYY-MM
    to get one specific historical month. Send 'start' and/or 'end' as YYYY-MM to get a series of months
    for charting or analysis; omitting both start and end while providing neither period returns the newest
    single report. Do not send 'period' together with 'start'/'end' — 'period' takes priority and a range
    will n
- name: curl_http_request
  title: Curl HTTP Request
  description: Fetch any URL from the cloud and return its full response.
  rest:
    method: POST
    server: https://curl-http-request.underscoredone.com
    path: /curl
    operationId: handler_curl_post
  input_schema_source: openapi/_original/done-curl-http-request-openapi.json#/components/schemas/Request
  inputs:
  - name: url
    type: string
    required: true
    description: The full web address you want to send a request to. Must start with http:// or https://.
  - name: method
    type: string
    required: false
    description: The type of request to make. Choose from GET, POST, PUT, PATCH, DELETE, or HEAD. Defaults
      to GET if not provided.
  price_usd: 0.01
  ai_instructions: Send a JSON body with a 'url' (required, must start with http:// or https://) and an
    optional 'method' field (defaults to GET). The response includes 'statusCode' (integer), 'headers'
    (object of key-value strings), 'body' (string, may be HTML, JSON, or plain text), 'durationMs' (how
    long the request took in milliseconds), and 'error' (null on success, or a plain-English message if
    the request coul
- name: directory_submission_lite
  title: Directory Submission Lite
  description: Orders submission of a website to twenty directories and returns a confirmation number.
  rest:
    method: POST
    server: https://directory-submission-lite.underscoredone.com
    path: /submit
    operationId: handler_submit_post
  input_schema_source: openapi/_original/done-directory-submission-lite-openapi.json#/components/schemas/Request
  inputs:
  - name: url
    type: string
    required: true
    description: The full web address of the site to submit, including the leading https://, for example
      https://example.com
  price_usd: 0.01
  ai_instructions: Call this once per website to order directory submissions. Save the submissionid from
    the reply — it cannot be recovered and it is the only way to retrieve the order. Do not call this
    again to check progress; that would place a second paid order. To check progress, send the confirmation
    number to the /status address on this same service, which is free.
- name: directory_submission_lite
  title: Directory Submission Lite
  description: Orders submission of a website to twenty directories and returns a confirmation number.
  rest:
    method: POST
    server: https://directory-submission-lite.underscoredone.com
    path: /status
    operationId: status_handler_status_post
  input_schema_source: openapi/_original/done-directory-submission-lite-openapi.json#/components/schemas/StatusRequest
  inputs:
  - name: submissionid
    type: string
    required: true
    description: The confirmation number returned when the order was placed
  price_usd: 0.01
  ai_instructions: Call this once per website to order directory submissions. Save the submissionid from
    the reply — it cannot be recovered and it is the only way to retrieve the order. Do not call this
    again to check progress; that would place a second paid order. To check progress, send the confirmation
    number to the /status address on this same service, which is free.
- name: dmarc_lookup
  title: Bulk DMARC Record Lookup
  description: Look up DMARC email security records for one or many domains at once.
  rest:
    method: POST
    server: https://dmarc-lookup.underscoredone.com
    path: /dmarc-lookup
    operationId: handler_dmarc_lookup_post
  input_schema_source: openapi/_original/done-dmarc-lookup-openapi.json#/components/schemas/Request
  inputs:
  - name: domains
    type: array
    required: true
    description: 'A list of one or more domain names to check. Use bare domain names only — no http://,
      https://, or www. Example: ["onescales.com", "hillms.com"]'
  price_usd: 0.01
  ai_instructions: 'Call this endpoint with POST and a JSON body containing a ''domains'' array of bare
    domain names (no http:// or www), e.g. {"domains": ["example.com", "mycompany.org"]}. Comma-separated
    strings inside an array entry are also tolerated and will be split. Each domain in the response will
    have: ''domain'' (the input domain), ''recordStatus'' (''Valid'', ''Invalid'', or ''Not Found''),
    ''domainPolicy'' (the p= tag '
- name: dns_whois_lookup
  title: DNS & WHOIS Lookup
  description: Get all DNS records and WHOIS data for up to 10 domains at once
  rest:
    method: POST
    server: https://dns-whois-lookup.underscoredone.com
    path: /lookup
    operationId: handler_lookup_post
  input_schema_source: openapi/_original/done-dns-whois-lookup-openapi.json#/components/schemas/Request
  inputs:
  - name: domains
    type: array
    required: true
    description: A list of domain names or full web addresses to look up. You can pass bare domains like
      'example.com', or full addresses like 'https://www.example.com/page' — the system will automatically
      extract just the domain. Send b
  price_usd: 0.01
  ai_instructions: POST an array of domain names (or full URLs) under the key 'domains'. You may send
    between 1 and 10 entries. The response is an array of objects, one per unique domain, each containing
    fields prefixed with 'dns_' (all 13 record types) and 'whois_' (full registration data). All fields
    are always present; empty string means no record found or data not available. Multiple values within
    a single recor
- name: domain_age_checker
  title: Domain Age Checker
  description: Check how old a list of domains is and when they expire
  rest:
    method: POST
    server: https://domain-age-checker.underscoredone.com
    path: /lookup
    operationId: handler_lookup_post
  input_schema_source: openapi/_original/done-domain-age-checker-openapi.json#/components/schemas/DomainLookupRequest
  inputs:
  - name: domains
    type: array
    required: true
    description: A list of domain names you want to look up. You can include plain domains like 'example.com'
      or full web addresses like 'https://example.com/page' — both work. Maximum 50 domains per request.
  price_usd: 0.01
  ai_instructions: 'POST a JSON body with a ''domains'' list of domain name strings (plain domains or
    full URLs both work). Each result includes ''domainName'', ''age'' as a human-readable string, ''registrationDate'',
    ''lastUpdateDate'', ''expirationDate'', ''expiringIn'' as a human-readable string, and ''status''
    which is ''success'' or ''error''. If WHOIS data cannot be retrieved for a domain, that entry will
    have status ''error'' and '
- name: domain_availability_checker
  title: Domain Availability Checker
  description: Look up whether domain names are registered using official RDAP registry data.
  rest:
    method: POST
    server: https://domain-availability-checker.underscoredone.com
    path: /check
    operationId: check_check_post
  input_schema_source: openapi/_original/done-domain-availability-checker-openapi.json#/components/schemas/CheckRequest
  inputs:
  - name: domains
    type: array
    required: true
    description: Between 1 and 10 domain names to check, like 'example.com'. Use the plain domain only
      — no 'https://', no 'www.', no subdomains.
  price_usd: 0.01
  ai_instructions: 'POST up to 10 plain registrable domain names (like ''example.com'', not ''https://example.com''
    or ''sub.example.com'') in the ''domains'' array. Each result has a ''registered'' field: true means
    a record exists, false means none was found, null means the result could not be determined. A false
    result does NOT guarantee the domain is purchasable — reserved and premium names can also return false.
    The ''tld_'
- name: ens_resolver
  title: ENS Resolver
  description: Look up an ENS name or address and get back the matching one.
  rest:
    method: POST
    server: https://ens-resolver.underscoredone.com
    path: /resolve
    operationId: handler_resolve_post
  input_schema_source: openapi/_original/done-ens-resolver-openapi.json#/components/schemas/Request
  inputs:
  - name: query
    type: string
    required: true
    description: An ENS name like vitalik.eth, or an Ethereum address starting with 0x. The service automatically
      figures out which one you gave it.
  price_usd: 0.01
  ai_instructions: 'Send the query field with either an ENS name ending in .eth or a 0x Ethereum address.
    The service auto-detects direction: names resolve forward to addresses, addresses resolve backward
    to their primary ENS name. If nothing is found, resolved will be false and address or name may be
    null, but this is not an error. Only malformed input causes a 400 error.'
- name: hackernews_data
  title: Hacker News Data
  description: Get current Hacker News lists or scan a past date for stories, jobs, and polls, with optional
    comment threads.
  rest:
    method: POST
    server: https://hackernews-data.underscoredone.com
    path: /fetch
    operationId: handler_fetch_post
  input_schema_source: openapi/_original/done-hackernews-data-openapi.json#/components/schemas/Request
  inputs:
  - name: category
    type: any
    required: false
    description: 'Which current list to pull from: top, new, best, ask, show, or job. Ignored if a date
      is given.'
  - name: date
    type: any
    required: false
    description: A calendar day in YYYY-MM-DD format (UTC). If given, the tool switches to scanning that
      whole day instead of using a category.
  - name: limit
    type: any
    required: false
    description: The largest number of results to return. Default is 30, maximum is 200.
  - name: comment_limit
    type: any
    required: false
    description: How many top-level comments to fetch per story. Default 10, maximum 100, 0 means none.
  - name: reply_depth
    type: any
    required: false
    description: How many levels of nested replies to fetch below each comment. Default 0, maximum 5.
  - name: replies_per_comment
    type: any
    required: false
    description: The largest number of replies fetched per comment at each level. Default 5, maximum 20.
  - name: page_token
    type: any
    required: false
    description: A resume point from a previous date-scan response, used to continue scanning the same
      day where it left off.
  price_usd: 0.01
  ai_instructions: 'Call this when you need Hacker News content: either a current ranked list (top, new,
    best, ask, show, job) or every story/job/poll posted on a given UTC date. Set ''date'' to switch to
    date mode; leave it empty to use category mode. Use ''reply_depth'' and ''comment_limit'' sparingly
    since higher values take longer. For date mode, if ''range_complete'' is false in the response, call
    again passing the retu'
- name: hash_hmac
  title: Hashing, HMAC & Checksum Suite
  description: Hash, sign, and verify data fingerprints in one call
  rest:
    method: POST
    server: https://hash-hmac.underscoredone.com
    path: /hash
    operationId: handler_hash_post
  input_schema_source: openapi/_original/done-hash-hmac-openapi.json#/components/schemas/Request
  inputs:
  - name: input
    type: string
    required: true
    description: The text or data you want to hash. For binary data, base64-encode it first and set input_encoding
      to 'base64'.
  - name: input_encoding
    type: string
    required: false
    description: How the input is encoded. Use 'utf-8' for normal text (default) or 'base64' for binary
      data.
  - name: algorithms
    type: array
    required: true
    description: 'Which fingerprint methods to use. Choose any of: md5, sha1, sha224, sha256, sha384,
      sha512, sha3_256, sha3_512, blake2b, blake2s, crc32, adler32. You may list several at once.'
  - name: output_encoding
    type: string
    required: false
    description: How to display the resulting fingerprints. Choose 'hex' (default), 'base64', or 'base64url'
      (safe for URLs).
  - name: hmac
    type: any
    required: false
    description: Optional. Include this block to sign data or verify a signature.
  price_usd: 0.01
  ai_instructions: POST to /hash with an 'input' string and an 'algorithms' list such as ['sha256','crc32'].
    To verify a webhook signature set hmac.mode to 'verify', provide hmac.key and hmac.expected_signature
    — the response field hmac.matches will be true or false. Never guess or generate hash output yourself;
    always call this endpoint. The same input always produces the same output. Use input_encoding 'base64'
    wh
- name: http_header_checker
  title: HTTP Header Checker
  description: Look up the response headers and status of any URL
  rest:
    method: POST
    server: https://http-header-checker.underscoredone.com
    path: /check
    operationId: handler_check_post
  input_schema_source: openapi/_original/done-http-header-checker-openapi.json#/components/schemas/Request
  inputs:
  - name: url
    type: string
    required: true
    description: The full web address to check, including http:// or https://
  - name: method
    type: any
    required: false
    description: Either GET or HEAD. GET fetches the page normally, HEAD only asks for headers without
      the body.
  - name: follow_redirects
    type: any
    required: false
    description: Whether to follow redirects to their final destination or stop at the first one.
  price_usd: 0.01
  ai_instructions: Call this when you need to inspect what a web server returns before downloading content,
    such as checking caching rules, security headers, redirect chains, or content types. Provide a full
    URL with http:// or https://. Do not use this to scrape page content, only headers and status metadata
    are returned.
- name: http_status_checker
  title: HTTP Status Code Checker
  description: Check live status and redirects for up to 10 URLs in one call.
  rest:
    method: POST
    server: https://http-status-checker.underscoredone.com
    path: /check
    operationId: check_urls_check_post
  input_schema_source: openapi/_original/done-http-status-checker-openapi.json#/components/schemas/CheckRequest
  inputs:
  - name: urls
    type: array
    required: true
    description: A list of 1 to 10 web addresses you want to check. Each address must start with http://
      or https://.
  price_usd: 0.01
  ai_instructions: 'POST a JSON body with a field called ''urls'' containing a list of 1 to 10 fully-qualified
    web addresses (must start with http:// or https://). Each URL is checked independently. The response
    is a list of result objects in the same order as the input. Each result includes the final status
    code seen, whether a redirect happened, and the destination URL if redirected. If a URL could not
    be reached at '
- name: json_suite
  title: JSON Validate, Repair, Convert, Query & Diff
  description: Validate, repair, format, query, diff, convert, and canonicalize JSON in one call.
  rest:
    method: POST
    server: https://json-suite.underscoredone.com
    path: /json
    operationId: handler_json_post
  input_schema_source: openapi/_original/done-json-suite-openapi.json#/components/schemas/Request
  inputs:
  - name: op
    type: string
    required: true
    description: 'What you want to do. Choose one of: validate, repair, format, canonicalize, convert,
      query, diff, patch, schema, stats.'
  - name: data
    type: any
    required: false
    description: The JSON value you want to work with — can be an object, array, string, number, or anything
      valid. Required for all operations except validate and repair.
  - name: data_raw
    type: any
    required: false
    description: The raw text you want to validate or repair. Use this instead of data only for validate
      and repair, because those need to see the original broken text exactly as-is.
  - name: options
    type: any
    required: false
    description: 'Extra settings for the chosen operation. For example: indent, sort_keys, jsonpath, pointer,
      to (for convert), schema, data_b (for diff/patch). All optional — sensible defaults are used.'
  price_usd: 0.01
  ai_instructions: 'Send the ''op'' field to choose what to do. Use ''data_raw'' (a plain string) only
    for ''validate'' and ''repair'' — these two need the original broken text. Use ''data'' (actual parsed
    JSON) for all other operations: format, canonicalize, convert, query, diff, patch, schema, stats.
    For ''diff'' and ''patch'', put the first document in ''data'' and the second in ''options.data_b''.
    For ''query'', put your JSONPath ex'
- name: leetspeak_translator
  title: Leetspeak Translator
  description: Encode text to leetspeak or decode leetspeak back to normal text
  rest:
    method: POST
    server: https://leetspeak-translator.underscoredone.com
    path: /translate
    operationId: handler_translate_post
  input_schema_source: openapi/_original/done-leetspeak-translator-openapi.json#/components/schemas/Request
  inputs:
  - name: text
    type: string
    required: true
    description: The words you want to convert, either everyday writing or leetspeak.
  - name: direction
    type: string
    required: true
    description: Choose 'to_leet' to turn normal words into leetspeak, or 'to_normal' to turn leetspeak
      back into everyday words.
  - name: level
    type: string
    required: false
    description: Only used when turning words into leetspeak. Choose 'basic', 'intermediate', or 'advanced'
      for how many letters get changed.
  price_usd: 0.01
  ai_instructions: Send the text to convert, pick a direction ('to_leet' or 'to_normal'), and optionally
    a level ('basic', 'intermediate', 'advanced') for how heavily to substitute characters when converting
    to leetspeak. The result is always the same for the same input, so it can be cached or reused safely.
- name: mortgage_amortization
  title: Mortgage & Loan Amortization Calculator
  description: Exact monthly payment and full amortization schedule for any fixed-rate loan
  rest:
    method: POST
    server: https://mortgage-amortization.underscoredone.com
    path: /calculate
    operationId: handler_calculate_post
  input_schema_source: openapi/_original/done-mortgage-amortization-openapi.json#/components/schemas/Request
  inputs:
  - name: principal
    type: number
    required: true
    description: The total amount borrowed, in dollars. Must be greater than zero.
  - name: annual_rate_percent
    type: number
    required: true
    description: The yearly interest rate as a plain percentage — enter 6.5 for 6.5%. Enter 0 for an interest-free
      loan. Must be zero or above.
  - name: term_months
    type: integer
    required: true
    description: How many monthly payments the loan spans. A 30-year mortgage is 360; a 5-year car loan
      is 60. Must be at least 1.
  - name: start_date
    type: string
    required: true
    description: The date of the very first payment, written as YYYY-MM-DD (for example, 2026-07-01).
  - name: extra_monthly_payment
    type: number
    required: false
    description: An optional extra amount you put toward principal every single month, on top of the required
      payment. Defaults to 0.
  - name: one_time_payments
    type: array
    required: false
    description: An optional list of one-time lump-sum principal payments. Each item needs a 'month' and
      an 'amount'.
  - name: summary_only
    type: boolean
    required: false
    description: Set to true if you only want the totals and scheduled payment — no row-by-row schedule.
  - name: rounding
    type: integer
    required: false
    description: How many decimal places to round every money value to. Default is 2 (cents). Valid range
      is 2 to 6.
  price_usd: 0.01
  ai_instructions: 'POST to /calculate with principal (positive number), annual_rate_percent (0 or above),
    term_months (1 or above), and start_date (YYYY-MM-DD for the first payment). Optional: extra_monthly_payment
    (recurring extra principal each month), one_time_payments (list of objects with ''month'' and ''amount''
    keys for lump-sum payments), summary_only (true to skip the row-by-row schedule array), and rounding
    (d'
- name: ocr
  title: OCR Text Extractor
  description: Pulls readable text out of a picture.
  rest:
    method: POST
    server: https://ocr.underscoredone.com
    path: /read
    operationId: handler_read_post
  input_schema_source: openapi/_original/done-ocr-openapi.json#/components/schemas/Request
  inputs:
  - name: image_url
    type: any
    required: false
    description: A public web address pointing to the picture. Leave this out if you are sending the picture
      as text instead.
  - name: image_base64
    type: any
    required: false
    description: The picture encoded as text. Leave this out if you are sending a web address instead.
  - name: lang
    type: any
    required: false
    description: The language of the words in the picture. Only English ('eng') is supported right now.
  price_usd: 0.01
  ai_instructions: Provide either image_url or image_base64, never both. Use this when you have a picture
    (screenshot, scanned document, receipt, sign) and need the text inside it as plain text. Check the
    confidence score before trusting the result for critical decisions; low confidence means the image
    may be blurry or low quality.
- name: port_scanner
  title: TCP Port Scanner
  description: Check which ports on a host are open and what services are listening on them
  rest:
    method: POST
    server: https://port-scanner.underscoredone.com
    path: /scan
    operationId: scan_scan_post
  input_schema_source: openapi/_original/done-port-scanner-openapi.json#/components/schemas/ScanRequest
  inputs:
  - name: target
    type: string
    required: true
    description: The hostname (like 'example.com') or numeric IP address (like '93.184.216.34') you want
      to scan. Must be a real, reachable address.
  - name: ports
    type: any
    required: false
    description: Which ports to check. Use a range like '1-1000', a comma-separated list like '22,80,443',
      or leave this out to automatically check the 1000 most commonly used ports.
  - name: scan_type
    type: any
    required: false
    description: How to probe each port. The only supported value right now is 'tcp_connect'. Defaults
      to 'tcp_connect' if left out.
  - name: timeout
    type: any
    required: false
    description: How many seconds the scan is allowed to run before giving up. Must be between 5 and 60.
      Defaults to 30 if left out.
  price_usd: 0.01
  ai_instructions: POST a JSON body with 'target' (hostname or IP as a string) and optionally 'ports'
    (a range like '1-1000' or a comma-separated list like '22,80,443'), 'scan_type' (only 'tcp_connect'
    is supported), and 'timeout' (whole number of seconds, 5 to 60). The response includes an 'open_ports'
    array — each item has 'port', 'protocol', 'state', 'service', and 'version'. Use 'ports_scanned' and
    'closed_or_fi
- name: qr_code_generator
  title: Bulk QR Code Generator
  description: Convert a list of texts into QR code PNG images in one request
  rest:
    method: POST
    server: https://qr-code-generator.underscoredone.com
    path: /generate
    operationId: generate_generate_post
  input_schema_source: openapi/_original/done-qr-code-generator-openapi.json#/components/schemas/QRRequest
  inputs:
  - name: texts
    type: array
    required: true
    description: The list of things you want turned into QR codes — website addresses, phone numbers,
      plain text, WiFi credentials, anything you like. Each entry must be 700 characters or fewer. Empty
      entries are skipped. At least one re
  - name: size
    type: any
    required: false
    description: How wide (and tall — QR codes are square) the output image should be, measured in pixels.
      Must be between 100 and 1200. Defaults to 600 if you leave it out.
  - name: error_correction
    type: any
    required: false
    description: How much damage or dirt the QR code can survive and still scan correctly. Choose L (least
      tough, smallest code), M, Q, or H (most tough, slightly larger code). Defaults to H, which is best
      for printed labels or stickers.
  price_usd: 0.01
  ai_instructions: 'POST a JSON body with a key ''texts'' containing an array of strings. Each string
    becomes one QR code. The response is an array of objects in the same order as the non-empty inputs,
    each containing ''text'' (the original string) and ''qr_base64'' (a data URI string starting with
    ''data:image/png;base64,'' ready for use in an HTML img src or file save). Strings longer than 700
    characters or that are empty '
- name: screenshots
  title: Screenshots
  description: Takes a screenshot of any webpage after it fully loads.
  rest:
    method: POST
    server: https://screenshots.underscoredone.com
    path: /capture
    operationId: handler_capture_post
  input_schema_source: openapi/_original/done-screenshots-openapi.json#/components/schemas/Request
  inputs:
  - name: url
    type: string
    required: true
    description: The web address of the page you want a picture of.
  - name: full_page
    type: boolean
    required: false
    description: If true, captures the entire scrollable page instead of just what is visible on screen.
  - name: viewport
    type: any
    required: false
    description: The width and height of the browser window used to show the page, in pixels.
  - name: wait_until
    type: string
    required: false
    description: When to consider the page ready to capture.
  - name: allow_popups
    type: boolean
    required: false
    description: If false, cookie banners, consent popups, and chat widgets are hidden before the picture
      is taken.
  - name: load_lazy
    type: boolean
    required: false
    description: If true, scrolls through the page first so slow-loading images finish loading before
      the picture is taken.
  price_usd: 0.01
  ai_instructions: Send the target url plus any optional capture settings. Set load_lazy true and wait_until
    to networkidle0 or networkidle2 for pages that load content as you scroll, such as image galleries
    or social feeds. Set allow_popups false (default) to automatically hide cookie banners and chat widgets.
    The response gives a short-lived link to the captured image, not the image itself, so download it
    promptly
- name: seo_data_extractor
  title: SEO Data Extractor
  description: Visit a URL and pull out every important SEO field in one call.
  rest:
    method: POST
    server: https://seo-data-extractor.underscoredone.com
    path: /extract
    operationId: handler_extract_post
  input_schema_source: openapi/_original/done-seo-data-extractor-openapi.json#/components/schemas/Request
  inputs:
  - name: url
    type: string
    required: true
    description: 'The full web address of the page you want to analyse, including http:// or https://.
      Example: https://www.example.com'
  - name: timeout_seconds
    type: integer
    required: false
    description: How many seconds to wait for the page to respond before giving up. Must be between 1
      and 15. Defaults to 10 if not provided.
  price_usd: 0.01
  ai_instructions: Send a single URL string in the 'url' field. The API will fetch that page and return
    structured SEO data. If the page redirects, the final destination URL and its status code are returned.
    All missing fields come back as empty strings, never null. The 'num_links' field is always an integer.
    Use the returned fields directly for SEO audits, content comparisons, or monitoring pipelines. Do
    not send m
- name: shopify_ai_rank_checker
  title: Shopify AI Rank Checker
  description: Finds your Shopify product's rank in the AI shopping catalog for any search keywords.
  rest:
    method: POST
    server: https://shopify-ai-rank-checker.underscoredone.com
    path: /rank
    operationId: handler_rank_post
  input_schema_source: openapi/_original/done-shopify-ai-rank-checker-openapi.json#/components/schemas/Request
  inputs:
  - name: shop_domain
    type: string
    required: true
    description: Your store's web address, for example legacybilliards.com or example-store.myshopify.com.
      You can paste it with or without https://.
  - name: keywords
    type: array
    required: true
    description: A list of 1 to 10 search phrases you want to check your store's rank for. Split larger
      lists across multiple calls.
  - name: report_detail
    type: string
    required: false
    description: Choose 'simple' for a quick rank summary or 'full' for complete competitor and data quality
      details.
  - name: ships_to_country
    type: string
    required: false
    description: A two-letter country code, such as US or DE, so results reflect what a shopper in that
      country would see.
  price_usd: 0.01
  ai_instructions: Call this when a user wants to know how visible their Shopify store is to AI shopping
    assistants for specific search terms. Provide the store's myshopify.com domain and a list of search
    keywords. Use 'simple' report_detail for a quick rank check, or 'full' for complete competitor data
    and a data quality score. Keep keyword lists to 10 or fewer per call for faster responses; split larger
    lists into
- name: sitemap_url_extractor
  title: Sitemap URL Extractor
  description: Extract all URLs from a sitemap.xml file, including nested sub-sitemaps, with metadata.
  rest:
    method: POST
    server: https://sitemap-url-extractor.underscoredone.com
    path: /extract
    operationId: handler_extract_post
  input_schema_source: openapi/_original/done-sitemap-url-extractor-openapi.json#/components/schemas/Request
  inputs:
  - name: sitemap_url
    type: string
    required: true
    description: 'The full web address of the sitemap.xml file you want to read, for example: https://example.com/sitemap.xml'
  price_usd: 0.01
  ai_instructions: POST a JSON body with a single field called 'sitemap_url' containing the full URL of
    the sitemap.xml file you want to parse. The API will follow any nested sitemap index files automatically
    and return a flat list of all discovered URLs along with their metadata. If the sitemap is unreachable,
    returns a 400 error. If the URL does not point to valid XML sitemap content, returns a 400 error.
    Do not p
- name: url_uptime_checker
  title: URL Uptime Checker
  description: Checks if a given web address is currently online and reachable.
  rest:
    method: POST
    server: https://url-uptime-checker.underscoredone.com
    path: /check
    operationId: handler_check_post
  input_schema_source: openapi/_original/done-url-uptime-checker-openapi.json#/components/schemas/Request
  inputs:
  - name: url
    type: string
    required: true
    description: The full web address to check, starting with http or https.
  - name: follow_redirects
    type: boolean
    required: false
    description: Whether to follow the site to its final address if it sends you somewhere else, before
      reporting the result.
  - name: method
    type: string
    required: false
    description: Which kind of check to run. GET reads the whole page. HEAD only checks that the site
      answers, which is quicker.
  price_usd: 0.01
  ai_instructions: Call this before treating any URL as reachable. Pass the full web address including
    http:// or https://. Read the 'up' field for a simple yes/no, 'status_code' for the response code,
    'response_time_ms' for speed, and 'error' for the reason if it failed. Do not retry automatically
    on timeout unless explicitly instructed.
cross_links:
  agentic_access: agentic-access/done-agentic-access.yml
  conventions: conventions/done-conventions.yml
  skills: skills/_index.yml
deployment:
  mode: none
  verified: derived
  tools: 27
  checked: '2026-08-12'
  source: catalog MCP census