MCP App Store
by KeenEthics
SigParser icon

SigParser MCP Server for ChatGPT

By SigParser

Overview

SigParser builds a database of contacts, companies, and relationship metrics from your team's email and calendar interactions. Use it in ChatGPT to find who you know at a company, see which coworker has the warmest relationship with a prospect, prepare pre-meeting company briefs, find dormant relationships worth re-engaging, export filtered contact lists to CSV, update contact and company records, and push contacts to your connected CRM or email marketing app.

How to connect

  1. ChatGPT: Open the SigParser listing in the ChatGPT app directory and choose Connect. ChatGPT listing SigParser exposes 14 tools here.

Tools

create_or_update_company

ChatGPT
Create a new company or update an existing company's fields. Provide the company data as JSON with at least a 'domain' field; only the fields you include are updated. Set describe_fields=true (with no companyData) to instead return the list of fields you can set — their names, types, descriptions, whether each is writable, and allowed values — so you can discover valid field names before writing. Returns the updated company data and which fields were created or updated.

create_or_update_contact

ChatGPT
Create a new contact or update an existing contact's fields. Provide the contact data as JSON with at least an 'email_address' field; only the fields you include are updated. Set describe_fields=true (with no contactData) to instead return the list of fields you can set — their names, types, descriptions, whether each is writable, and allowed values — so you can discover valid field names before writing. Returns the updated contact data and which fields were created or updated.

enrich_contact

ChatGPT
Run a third-party enrichment action against an existing contact (identified by email). Choose the action: 'job_enrichment' pulls current and past job experiences from the contact's LinkedIn profile (costs 10 credits) and returns the match status; 'email_validation' checks whether the email address is deliverable (small cost) and returns the validation status (Valid, Invalid, or Catch-all) plus sub-status. Both require sufficient credits. To read a stored result afterward, use query_contacts (filtering on email_address).

explain_field

ChatGPT
Explains why a contact, company, or relationship field has its current value. Returns the source of the value (email signature, user input, field rule, connector sync, etc.), the matched rule conditions if the value comes from a field rule, and the full history of observed values for the field. entity_type values: 'contact', 'company', 'relationship' - For contact/company: entity_id is the record's GUID — the 'record_id' field returned by query_contacts / query_accounts (backend name enterprisecontactid / enterpriseaccountid). - For relationship: entity_id is the composite id exposed by query_relationships — 'contactEmail&coworkerEmail' (the 'id' field). - field: the field name — either the backend name (e.g. 'title') or the iPaaS API name (e.g. 'job_title'). For relationships only the 'relationship_score' field is supported. All other relationship fields return explanation_type 'not_supported'.

export_contacts_to_connector

ChatGPT
Create or update one or more Contact records in a CRM like Salesforce, Dynamics, or HubSpot, or an email marketing app like MailChimp or Constant Contact. Address book connectors (Google Contacts, Outlook Contacts, Exchange Contacts) are not supported by this tool. If you don't specify the connector type or name it will return a list of connectors for you to choose from. The match on connector_type and connector_name is case-insensitive. If the user specifies a CRM type with multiple matching connectors, the primary CRM is selected automatically; otherwise the tool returns the list of matching connectors and asks you to pick one by name (or by source_id via connector_sourceid). The export runs as a background job. The tool waits up to 15 seconds for the job to complete and then returns the current status. To check progress on a job that took longer than 15 seconds, call this tool again with just the export_job_id from a previous response — it returns that job's current status and per-contact results instead of starting a new export.

get_download

ChatGPT
Retrieve a file that a query generated asynchronously. When you call a query_* tool with output_format=csv, it returns a download_id; pass that id here to check the file's status and get its download link. Returns a status: - 'queued' / 'in_progress' — the file is still being generated; call again to poll. - 'ready' — 'download_url' holds a link valid for about 30 minutes; give it to the user promptly. - 'failed' / 'abandoned' — it did not complete; request the CSV again from the query tool.

get_sigparser_skill

ChatGPT
IMPORTANT: Call this skill tool FIRST before using any other SigParser tools. Returns the SigParser relationship intelligence skill guide, which explains how to generate pre-meeting company briefs, follow-up recommendations for stale relationships, and industry network analysis reports using SigParser data, and how to export Contacts to external connectors (CRMs and email marketing apps). The response also includes the current list of connectors configured on the user's account so the caller has them in context.

get_users

ChatGPT
Get SigParser users. By default (scope="me") returns the current user — the one the API key belongs to — including their profile, role, and permissions. Set scope="team" to instead list the other users on the same team/enterprise (paginated, with an optional name/email search). Most callers want the default "me".

query_accounts

ChatGPT
Query accounts (companies) using structured filters. Good fields to select are: company_name, email_domain, company_website, company_employees_range, industry_name, interactions_total (total interaction count), interactions_total_latest (date of last interaction). Common filters: - By industry: {"field":"industry_name","operator":"eq","value":["Technology"]} - By domain: {"field":"domain","operator":"eq","value":["example.com"]} - By email volume: {"field":"stat_emails_v2","operator":"gt","value":["50"]} - By status: {"field":"status","operator":"eq","value":["Valid"]} Filter structure: outer array = OR groups, inner array = AND conditions within each group. - AND: [[{A},{B}]] — both conditions must match - OR: [[{A}],[{B}]] — either condition matches - OR of ANDs: [[{A},{B}],[{C},{D}]] — (A AND B) OR (C AND D) Operators are consistent across field types. Call the schema tool for the exact operators valid on each field. Some fields include a 'description' in the schema with usage notes for a special operator (e.g. the location fields' 'map' filter finds records within a distance of a place). - Equality: 'eq' / 'ne'. The value is an array, so multiple values act as IN: {"operator":"eq","value":["C-Level","Director","VP"]}. This works the same on status/select fields. - Numbers: 'gt' / 'lt'. For a range, combine 'gt' and 'lt' filters in the same group. - Text: 'contains', 'not_contains', 'starts_with', 'ends_with', 'contains_word', 'regex'. - Presence: 'is_null' / 'is_not_null' (has no value / has a value) — pass an empty value array. Available on most field types (text, number, date, location, ...) but not all — a status/select field, for example, always has a value; check the schema for the field. - Dates: 'after', 'before', 'last_days', 'before_days' (plus 'is_null' / 'is_not_null'). Dates use ISO 8601 format: "2026-01-15" Fields can be referenced by either their backend name (e.g. 'status') or their iPaaS API name (e.g. 'record_status') in filters, sort, and select. Use the schema tool to discover available fields. Note: response fields always use the iPaaS API name regardless of which name you pass in filters, sort, or select. When no 'select' is specified, the response returns a curated set of default fields to reduce payload size. Use 'select' to explicitly request additional fields if needed.

query_contacts

ChatGPT
Query contacts (people) using structured filters. Good fields to select are: email_address, record_status, job_title, name_full, phone_mobile, phone_work, company_name, interactions_total (total interaction count), interactions_total_latest (date of last interaction). Common filters: - By status: {"field":"status","operator":"eq","value":["Valid"]} - By title level: {"field":"job_level","operator":"eq","value":["C-Level","Director","VP"]} - By company: {"field":"domain","operator":"eq","value":["example.com"]} - By email activity: {"field":"stat_emails_v2","operator":"gt","value":["10"]} Filter structure: outer array = OR groups, inner array = AND conditions within each group. - AND: [[{A},{B}]] — both conditions must match - OR: [[{A}],[{B}]] — either condition matches - OR of ANDs: [[{A},{B}],[{C},{D}]] — (A AND B) OR (C AND D) Operators are consistent across field types. Call the schema tool for the exact operators valid on each field. Some fields include a 'description' in the schema with usage notes for a special operator (e.g. the location fields' 'map' filter finds records within a distance of a place). - Equality: 'eq' / 'ne'. The value is an array, so multiple values act as IN: {"operator":"eq","value":["C-Level","Director","VP"]}. This works the same on status/select fields. - Numbers: 'gt' / 'lt'. For a range, combine 'gt' and 'lt' filters in the same group. - Text: 'contains', 'not_contains', 'starts_with', 'ends_with', 'contains_word', 'regex'. - Presence: 'is_null' / 'is_not_null' (has no value / has a value) — pass an empty value array. Available on most field types (text, number, date, location, ...) but not all — a status/select field, for example, always has a value; check the schema for the field. - Dates: 'after', 'before', 'last_days', 'before_days' (plus 'is_null' / 'is_not_null'). Dates use ISO 8601 format: "2026-01-15" Fields can be referenced by either their backend name (e.g. 'status') or their iPaaS API name (e.g. 'record_status') in filters, sort, and select. Use the schema tool to discover available fields. Note: response fields always use the iPaaS API name regardless of which name you pass in filters, sort, or select. When no 'select' is specified, the response returns a curated set of default fields to reduce payload size. Use 'select' to explicitly request additional fields if needed.

query_coworkers

ChatGPT
Query coworkers (internal team members whose mailboxes are connected). Good fields to select are: employment_status, name, emailaddress, title. Common filters: - By activity: {"field":"stat_emails_v2","operator":"gt","value":["100"]} Filter structure: outer array = OR groups, inner array = AND conditions within each group. - AND: [[{A},{B}]] — both conditions must match - OR: [[{A}],[{B}]] — either condition matches - OR of ANDs: [[{A},{B}],[{C},{D}]] — (A AND B) OR (C AND D) Operators are consistent across field types. Call the schema tool for the exact operators valid on each field. Some fields include a 'description' in the schema with usage notes for a special operator (e.g. the location fields' 'map' filter finds records within a distance of a place). - Equality: 'eq' / 'ne'. The value is an array, so multiple values act as IN: {"operator":"eq","value":["C-Level","Director","VP"]}. This works the same on status/select fields. - Numbers: 'gt' / 'lt'. For a range, combine 'gt' and 'lt' filters in the same group. - Text: 'contains', 'not_contains', 'starts_with', 'ends_with', 'contains_word', 'regex'. - Presence: 'is_null' / 'is_not_null' (has no value / has a value) — pass an empty value array. Available on most field types (text, number, date, location, ...) but not all — a status/select field, for example, always has a value; check the schema for the field. - Dates: 'after', 'before', 'last_days', 'before_days' (plus 'is_null' / 'is_not_null'). Dates use ISO 8601 format: "2026-01-15" Fields can be referenced by either their backend name (e.g. 'status') or their iPaaS API name (e.g. 'record_status') in filters, sort, and select. Use the schema tool to discover available fields. Note: response fields always use the iPaaS API name regardless of which name you pass in filters, sort, or select. When no 'select' is specified, the response returns a curated set of default fields to reduce payload size. Use 'select' to explicitly request additional fields if needed.

query_job_experiences

ChatGPT
Query contact job experiences (historical job records). Each record represents a job a contact has held, including company, title, start/end dates, and whether it's their current role. Use this to track career history and job changes for contacts. Good fields to select are: name, company_name, company_domain, job_title, job_type, job_start_date, job_end_date. Common filters: - Current jobs only (no end date): {"field":"job_end_date","operator":"is_null","value":[]} - By company domain: {"field":"company_domain","operator":"eq","value":["example.com"]} - By title: {"field":"job_title","operator":"contains","value":["Engineer"]} Filter structure: outer array = OR groups, inner array = AND conditions within each group. - AND: [[{A},{B}]] — both conditions must match - OR: [[{A}],[{B}]] — either condition matches - OR of ANDs: [[{A},{B}],[{C},{D}]] — (A AND B) OR (C AND D) Operators are consistent across field types. Call the schema tool for the exact operators valid on each field. Some fields include a 'description' in the schema with usage notes for a special operator (e.g. the location fields' 'map' filter finds records within a distance of a place). - Equality: 'eq' / 'ne'. The value is an array, so multiple values act as IN: {"operator":"eq","value":["C-Level","Director","VP"]}. This works the same on status/select fields. - Numbers: 'gt' / 'lt'. For a range, combine 'gt' and 'lt' filters in the same group. - Text: 'contains', 'not_contains', 'starts_with', 'ends_with', 'contains_word', 'regex'. - Presence: 'is_null' / 'is_not_null' (has no value / has a value) — pass an empty value array. Available on most field types (text, number, date, location, ...) but not all — a status/select field, for example, always has a value; check the schema for the field. - Dates: 'after', 'before', 'last_days', 'before_days' (plus 'is_null' / 'is_not_null'). Dates use ISO 8601 format: "2026-01-15" Fields can be referenced by either their backend name (e.g. 'status') or their iPaaS API name (e.g. 'record_status') in filters, sort, and select. Use the schema tool to discover available fields. Note: response fields always use the iPaaS API name regardless of which name you pass in filters, sort, or select. When no 'select' is specified, the response returns a curated set of default fields to reduce payload size. Use 'select' to explicitly request additional fields if needed.

query_relationships

ChatGPT
Query coworker-to-contact relationships. Each record represents the interaction history between one coworker and one external contact, including email counts, last interaction dates, and meeting stats. Use this to find who on your team has the strongest relationship with a specific contact or company. Good fields to select are: coworker_emailaddress, coworker_name, contact_work_email, contact_name, account_name, total_interactions, emails_to, emails_from, total_meetings, latest_interaction. When filtering by an email address (contact_work_email or coworker_emailaddress), prefer the 'eq' operator over 'contains'. 'eq' uses an indexed exact match while 'contains' triggers a full table scan and is much slower — only fall back to 'contains' when the user genuinely needs partial-string matching. Default sort when omitted: total_interactions desc (strongest relationships first). Common filters: - By contact email: {"field":"contact_work_email","operator":"eq","value":["person@example.com"]} - By coworker email: {"field":"coworker_emailaddress","operator":"eq","value":["teammate@yourcompany.com"]} - By interaction volume: {"field":"total_interactions","operator":"gt","value":["10"]} Filter structure: outer array = OR groups, inner array = AND conditions within each group. - AND: [[{A},{B}]] — both conditions must match - OR: [[{A}],[{B}]] — either condition matches - OR of ANDs: [[{A},{B}],[{C},{D}]] — (A AND B) OR (C AND D) Operators are consistent across field types. Call the schema tool for the exact operators valid on each field. Some fields include a 'description' in the schema with usage notes for a special operator (e.g. the location fields' 'map' filter finds records within a distance of a place). - Equality: 'eq' / 'ne'. The value is an array, so multiple values act as IN: {"operator":"eq","value":["C-Level","Director","VP"]}. This works the same on status/select fields. - Numbers: 'gt' / 'lt'. For a range, combine 'gt' and 'lt' filters in the same group. - Text: 'contains', 'not_contains', 'starts_with', 'ends_with', 'contains_word', 'regex'. - Presence: 'is_null' / 'is_not_null' (has no value / has a value) — pass an empty value array. Available on most field types (text, number, date, location, ...) but not all — a status/select field, for example, always has a value; check the schema for the field. - Dates: 'after', 'before', 'last_days', 'before_days' (plus 'is_null' / 'is_not_null'). Dates use ISO 8601 format: "2026-01-15" Fields can be referenced by either their backend name (e.g. 'status') or their iPaaS API name (e.g. 'record_status') in filters, sort, and select. Use the schema tool to discover available fields. Note: response fields always use the iPaaS API name regardless of which name you pass in filters, sort, or select. When no 'select' is specified, the response returns a curated set of default fields to reduce payload size. Use 'select' to explicitly request additional fields if needed.

query_schema

ChatGPT
Get the schema for one grid — its available fields, their data types, and supported filter operators. Call this before the matching query_<grid> tool to discover what fields and operators are available. Set 'grid' to one of: contacts, accounts (or companies), coworkers, relationships, job_experiences. Each field shows both its backend name and its iPaaS API name (api_name). Either name can be used in the query tool's filters, sort, and select. IMPORTANT: Use the 'search' parameter to filter fields by keyword or group name. The full schema is very large and may exceed response limits. Always search rather than requesting the full schema. Search by group name to get all related fields at once (e.g. 'Interactions', 'Email Address', 'Location', 'Company', 'Job Title'). The response includes an available_groups list showing all groups for the grid.

Capabilities

Writes

Example Prompts

Click any prompt to copy it.

Similar Apps

Listed on MCP App Store

Listed on MCP App Store

Add this badge to your site or README. It links to this page.

HTML

<a href="https://mcp-app-store.keenethics-labs.com/app/sigparser"><img src="https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg" alt="Listed on MCP App Store" width="220" height="44"></a>

Markdown

[![Listed on MCP App Store](https://mcp-app-store.keenethics-labs.com/badges/listed-on-mcp-app-store.svg)](https://mcp-app-store.keenethics-labs.com/app/sigparser)