pm-gong

name	pm-gong
description	Gong connector knowledge and safe action guide.

pm-gong

Purpose

Reads Gong users, calls, scorecards, settings, flows, and related public API resources; executes selected typed POST read-query commands; models Gong mutations, multipart uploads, and top-level array uploads as typed reverse-ETL actions.

Icon

• asset: icons/gong.svg
• source: upstream_registry
• review_status: upstream_seeded
• review_url: https://us-66463.app.gong.io/settings/api/documentation

Capabilities

• check=true catalog=true read=true write=true query=false
• Integration type: api

Authentication

• Use pm credentials add with --from-env or --value-stdin for secret fields.

Configuration

• base_url
• max_pages
• mode
• page_size
• start_date
• access_key (secret)
• access_key_secret (secret)

ETL Streams

• users:
  • primary key: id
  • cursor: created
  • fields: active(), created(), email_address(), first_name(), id(), last_name(), manager_id(), phone_number(), title()
• calls:
  • primary key: id
  • cursor: started
  • fields: direction(), duration(), id(), is_private(), language(), media(), scheduled(), scope(), started(), system(), title(), url()
• scorecards:
  • primary key: scorecardId
  • cursor: updated
  • fields: created(), enabled(), scorecardId(), scorecardName(), updated(), workspaceId()
• crm_integrations:
  • fields: created(), id(), name(), title(), updated()
• workspaces:
  • fields: created(), id(), name(), title(), updated()
• trackers:
  • fields: created(), id(), name(), title(), updated()
• briefs:
  • fields: created(), id(), name(), title(), updated()
• library_folders:
  • fields: created(), id(), name(), title(), updated()
• flows:
  • fields: created(), id(), name(), title(), updated()
• flow_folders:
  • fields: created(), id(), name(), title(), updated()
• call_outcomes:
  • fields: created(), id(), name(), title(), updated()
• permission_profiles:
  • fields: created(), id(), name(), title(), updated()

Sync Modes

• ETL sync modes: full_refresh_append, full_refresh_overwrite, full_refresh_overwrite_deduped, incremental_append, incremental_append_deduped

Reverse ETL Actions

• add_call:
  • endpoint: POST /calls
  • required fields: actualStart, clientUniqueId, direction, parties, primaryUser
  • risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval
• create_permission_profile:
  • endpoint: POST /permission-profile?workspaceId={{ record.workspaceId }}
  • required fields: workspaceId
  • risk: high: administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation
• update_permission_profile:
  • endpoint: PUT /permission-profile?profileId={{ record.profileId }}
  • required fields: profileId
  • risk: high: administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation
• update_meeting:
  • endpoint: PUT /meetings/{meetingId}
  • required fields: meetingId, endTime, invitees, organizerEmail, startTime
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• delete_meeting:
  • endpoint: DELETE /meetings/{meetingId}
  • required fields: meetingId
  • risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation
• content_viewed:
  • endpoint: PUT /customer-engagement/content/viewed
  • required fields: contentId, contentTitle, contentUrl, eventTimestamp, reportingSystem
  • risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval
• content_shared:
  • endpoint: PUT /customer-engagement/content/shared
  • required fields: contentId, contentTitle, contentUrl, eventTimestamp, reportingSystem
  • risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval
• custom_action:
  • endpoint: PUT /customer-engagement/action
  • required fields: eventTimestamp, reportingSystem
  • risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval
• register_crm_integration:
  • endpoint: PUT /crm/integrations
  • required fields: name, ownerEmail
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• delete_crm_integration:
  • endpoint: DELETE /crm/integrations?integrationId={{ record.integrationId }}&clientRequestId={{ record.clientRequestId }}
  • required fields: integrationId, clientRequestId
  • risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation
• add_calls_users_access:
  • endpoint: PUT /calls/users-access
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• delete_calls_users_access:
  • endpoint: DELETE /calls/users-access
  • risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation
• create_meeting:
  • endpoint: POST /meetings
  • required fields: endTime, invitees, organizerEmail, startTime
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• integration_settings:
  • endpoint: POST /integration-settings
  • required fields: integrationTypeSettings
  • risk: high: administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation
• unassign_flows_by_instance_id:
  • endpoint: POST /flows/prospects/unassign-flows-by-instance-id
  • required fields: flowInstanceIds
  • risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation
• unassign_flows_by_crm_id:
  • endpoint: POST /flows/prospects/unassign-flows-by-crm-id
  • required fields: crmProspectId
  • risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation
• submit_flow_prospects_bulk_assignment:
  • endpoint: POST /flows/prospects/bulk-assignments
  • required fields: flowId, flowInstanceOwnerEmail, prospects
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• assign_prospects:
  • endpoint: POST /flows/prospects/assign
  • required fields: crmProspectsIds, flowId, flowInstanceOwnerEmail
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• assign_prospects_cool_off_override:
  • endpoint: POST /flows/prospects/assign/cool-off-override
  • required fields: crmProspectsIds, flowId, flowInstanceOwnerEmail
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• add_digital_interaction:
  • endpoint: POST /digital-interaction
  • required fields: content, eventId, eventType, timestamp
  • risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval
• purge_phone_number:
  • endpoint: POST /data-privacy/erase-data-for-phone-number?phoneNumber={{ record.phoneNumber }}
  • required fields: phoneNumber
  • risk: critical: destructive Gong data privacy erasure; requires reverse ETL plan, preview, approval, and destructive confirmation
• purge_email_address:
  • endpoint: POST /data-privacy/erase-data-for-email-address?emailAddress={{ record.emailAddress }}
  • required fields: emailAddress
  • risk: critical: destructive Gong data privacy erasure; requires reverse ETL plan, preview, approval, and destructive confirmation
• update_task:
  • endpoint: PATCH /tasks/{taskId}
  • required fields: taskId, userId
  • risk: medium: mutates Gong API state; requires reverse ETL approval
• upload_call_media:
  • endpoint: PUT /v2/calls/{{ record.id }}/media
  • required fields: id, media_file_path
  • risk: high: uploads call media from a bounded local file path to Gong; file path/content are redacted in plans and require reverse ETL approval
• upload_crm_entities:
  • endpoint: POST /v2/crm/entities
  • required fields: data_file_path
  • risk: high: uploads CRM data from a bounded local file path to Gong; file path/content are redacted in plans and require reverse ETL approval
• upload_crm_entity_schema:
  • endpoint: POST /v2/crm/entity-schema
  • required fields: selected_fields
  • risk: high: uploads CRM entity schema as a bounded top-level JSON array; use only reviewed table mappings and reverse ETL approval

Security

• read risk: external Gong API read of call, user, CRM, settings, flow, and activity data; direct reads are bounded and redacted
• write risk: typed Gong reverse ETL mutations for calls, meetings, CRM, permissions, flows, engagement, multipart uploads, top-level array CRM schema upload, and data privacy erasure
• approval: reverse ETL writes require plan, preview, approval, execute; destructive/admin actions require --confirm destructive; local file upload plans bind a SHA-256 payload identity that is snapshot-verified before network send
• Never pass secret values in chat, shell arguments, logs, docs, or JSON output.

Command Surface

• Inspect, read, and safely plan typed Gong operations.
• Usage: pm gong [flags]
• Source CLI: Gong API (Public OpenAPI 3.0.1)
• Global flags:
  • --credential (string): Credential name to use for the Gong request.
  • --connection (string): Alias for --credential.
  • --config (string_array): Connector config override as key=value.
  • --json (boolean): Emit machine-readable JSON output.
  • --limit (integer): Maximum ETL records to emit for stream commands.
  • --max-bytes (integer): Maximum direct-read response bytes; typed operations are capped at 16 MiB and each operation may declare a lower limit.
  • --plan (string): Execute an approved reverse-ETL plan by id.
  • --preview (boolean): Preview a reverse-ETL write command without making a network mutation.
  • --approve (string): Approval token required to execute a reverse-ETL plan.
  • --confirm (string): Typed confirmation challenge for destructive reverse-ETL writes.
• Calls
  • calls list - List Gong calls as ETL records. [intent=etl availability=implemented stream=calls]
  • calls get - Retrieve data for a specific call (/v2/calls/{id}) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --id
  • calls create - sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval [intent=reverse_etl availability=partial write=add_call]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --actualStart, --clientUniqueId, --direction, --parties, --primaryUser
  • calls users-access add - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=add_calls_users_access]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.
  • calls users-access delete - removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=delete_calls_users_access]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.
  • calls extensive - Retrieve detailed call data by various filters (/v2/calls/extensive) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --call-id, --from, --to, --primary-user-id, --workspace-id, --cursor, --context, --context-timing, --include-parties, --include-media, --include-public-comments, --include-brief, --include-call-outcome, --include-highlights, --include-key-points, --include-outline, --include-structure, --include-topics, --include-trackers, --include-tracker-occurrences, --include-speakers, --include-questions, --include-person-stats, --include-video-stats
  • calls users-access get - Retrieve users that have specific individual access to calls (/v2/calls/users-access) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --call-id
  • calls transcript - Retrieve transcripts of calls (/v2/calls/transcript) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --call-id, --from, --to, --workspace-id, --cursor
  • calls upload-media - Add call media (/v2/calls/{id}/media) [intent=reverse_etl availability=implemented write=upload_call_media]; approval: reverse ETL plan -> preview -> approval -> execute; destructive confirmation required; risk: high: uploads call media from a bounded project-local file path; file path/content are redacted in plans; notes: Uses typed multipart write support; no generic upload command is exposed.; flags: --id, --media-file-path
• Users
  • users list - List Gong users as ETL records. [intent=etl availability=implemented stream=users]
  • users get - Retrieve user (/v2/users/{id}) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --id
  • users settings-history - Retrieve user settings history (/v2/users/{id}/settings-history) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --id
  • users extensive - List users by filter (/v2/users/extensive) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --user-id, --created-from, --created-to, --include-avatars, --cursor
• Scorecards
  • scorecards list - List Gong scorecards as ETL records. [intent=etl availability=implemented stream=scorecards]
• Workspaces
  • workspaces list - List Gong workspaces as ETL records. [intent=etl availability=implemented stream=workspaces]
• Settings
  • settings trackers list - List Gong tracker settings as ETL records. [intent=etl availability=implemented stream=trackers]
  • settings briefs list - List Gong AI brief settings as ETL records. [intent=etl availability=implemented stream=briefs]
• Permissions
  • permissions profiles list - List Gong permission profiles as ETL records. [intent=etl availability=implemented stream=permission_profiles]
  • permissions profile get - Permission profile for a given profile Id (/v2/permission-profile) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --profileId
  • permissions profile-users list - List all users attached to a given permission profile (/v2/permission-profile/users) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --profileId
  • permissions profile create - administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=create_permission_profile]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --workspaceId
  • permissions profile update - administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=update_permission_profile]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --profileId
• CRM
  • crm integrations list - List Generic CRM integrations as ETL records. [intent=etl availability=implemented stream=crm_integrations]
  • crm entity-schema get - List Schema Fields (/v2/crm/entity-schema) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --integrationId, --objectType
  • crm entities get - Get CRM objects (/v2/crm/entities) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --integrationId, --objectType, --objectsCrmIds
  • crm request-status get - Get Request Status (/v2/crm/request-status) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --integrationId, --clientRequestId
  • crm integrations register - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=register_crm_integration]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --name, --ownerEmail
  • crm integrations delete - removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=delete_crm_integration]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --clientRequestId, --integrationId
  • crm upload-entities - Upload CRM objects (/v2/crm/entities) [intent=reverse_etl availability=implemented write=upload_crm_entities]; approval: reverse ETL plan -> preview -> approval -> execute; destructive confirmation required; risk: high: uploads CRM data from a bounded project-local file path; file path/content are redacted in plans; notes: Uses typed multipart write support; no generic upload command is exposed.; flags: --data-file-path
  • crm upload-entity-schema - Upload Object Schema (/v2/crm/entity-schema) [intent=reverse_etl availability=partial write=upload_crm_entity_schema]; approval: reverse ETL plan -> preview -> approval -> execute; destructive confirmation required; risk: high: uploads CRM entity schema as a top-level JSON array; no raw CLI body is exposed; notes: Write action supports schema-gated top-level array bodies for table/reverse-ETL mappings; provider-style raw JSON CLI flags remain intentionally unavailable.
• Meetings
  • meetings update - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=update_meeting]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --endTime, --invitees, --meetingId, --organizerEmail, --startTime
  • meetings delete - removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=delete_meeting]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --meetingId
  • meetings create - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=create_meeting]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --endTime, --invitees, --organizerEmail, --startTime
  • meetings integration-status - Validate Gong meeting Integration (/v2/meetings/integration/status) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded POST read-query; sends only typed email array body fields and redacts sensitive response fields; notes: Executes through the typed operation direct-read engine; no raw body flag is exposed.; flags: --email
• Engagement
  • engagement content-viewed - sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval [intent=reverse_etl availability=partial write=content_viewed]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --contentId, --contentTitle, --contentUrl, --eventTimestamp, --reportingSystem
  • engagement content-shared - sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval [intent=reverse_etl availability=partial write=content_shared]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --contentId, --contentTitle, --contentUrl, --eventTimestamp, --reportingSystem
  • engagement custom-action - sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval [intent=reverse_etl availability=partial write=custom_action]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --eventTimestamp, --reportingSystem
• Stats
  • stats interaction - Retrieve interaction stats for applicable users by date (/v2/stats/interaction) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --from-date, --to-date, --user-id, --created-from, --created-to, --cursor
  • stats activity-scorecards - Retrieve answered scorecards for applicable reviewed users or scorecards for a date range (/v2/stats/activity/scorecards) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --call-from-date, --call-to-date, --review-from-date, --review-to-date, --review-method, --reviewed-user-id, --scorecard-id, --cursor
  • stats activity-day-by-day - Retrieve daily activity for applicable users for a date range (/v2/stats/activity/day-by-day) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --from-date, --to-date, --user-id, --created-from, --created-to, --cursor
  • stats activity-aggregate - Retrieve aggregated activity for defined users by date (/v2/stats/activity/aggregate) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --from-date, --to-date, --user-id, --created-from, --created-to, --cursor
  • stats activity-aggregate-by-period - Retrieve aggregated activity for defined users by a date range with grouping in time periods (/v2/stats/activity/aggregate-by-period) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --from-date, --to-date, --user-id, --created-from, --created-to, --cursor, --aggregation-period
• Engage Flows
  • flows list - List Gong Engage flows as ETL records. [intent=etl availability=implemented stream=flows]
  • flows folders list - List Gong Engage flow folders as ETL records. [intent=etl availability=implemented stream=flow_folders]
  • flows bulk-assignment get - Get the results of a bulk assignment of prospects to a flow (/v2/flows/prospects/bulk-assignments/{id}) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --id
  • flows prospects unassign-by-instance - removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=unassign_flows_by_instance_id]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --flowInstanceIds
  • flows prospects unassign-by-crm - removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=unassign_flows_by_crm_id]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: removes Gong access, integration, meeting, or flow assignment state; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --crmProspectId
  • flows prospects bulk-assign - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=submit_flow_prospects_bulk_assignment]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --flowId, --flowInstanceOwnerEmail, --prospects
  • flows prospects assign - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=assign_prospects]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --crmProspectsIds, --flowId, --flowInstanceOwnerEmail
  • flows prospects assign-cool-off-override - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=assign_prospects_cool_off_override]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --crmProspectsIds, --flowId, --flowInstanceOwnerEmail
  • flows steps - Get flow details and steps for one or more Engage flows (/v2/flows/steps) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded POST read-query; sends only typed flow ID array body fields and redacts sensitive response fields; notes: Executes through the typed operation direct-read engine; no raw body flag is exposed.; flags: --flow-id
  • flows prospects - List prospects assigned flows (/v2/flows/prospects) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded POST read-query; sends only typed CRM prospect or flow instance ID arrays and redacts sensitive response fields; notes: Executes through the typed operation direct-read engine; no raw body flag is exposed.; flags: --crm-prospect-id, --flow-instance-id
• Library
  • library folders list - List Gong library folders as ETL records. [intent=etl availability=implemented stream=library_folders]
  • library folder-content list - List of calls in a specific folder (/v2/library/folder-content) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --folderId
• Entities
  • entities get-brief - Generate a brief for a CRM entity (/v2/entities/get-brief) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --workspaceId, --briefName, --crmEntityType, --crmEntityId, --timePeriod, --fromDateTime, --toDateTime
  • entities ask - Ask about an entity (/v2/entities/ask-entity) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --workspaceId, --crmEntityType, --crmEntityId, --timePeriod, --fromDateTime, --toDateTime, --question
• Data Privacy
  • privacy find-phone - Retrieve all references to a phone number. (/v2/data-privacy/data-for-phone-number) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --phoneNumber
  • privacy find-email - Retrieve all references to an email address. (/v2/data-privacy/data-for-email-address) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --emailAddress
  • privacy erase-phone - destructive Gong data privacy erasure; requires reverse ETL plan, preview, approval, and destructive confirmation [intent=reverse_etl availability=partial write=purge_phone_number]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: critical: destructive Gong data privacy erasure; requires reverse ETL plan, preview, approval, and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --phoneNumber
  • privacy erase-email - destructive Gong data privacy erasure; requires reverse ETL plan, preview, approval, and destructive confirmation [intent=reverse_etl availability=partial write=purge_email_address]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: critical: destructive Gong data privacy erasure; requires reverse ETL plan, preview, approval, and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --emailAddress
• Tasks
  • tasks update - mutates Gong API state; requires reverse ETL approval [intent=reverse_etl availability=partial write=update_task]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: medium: mutates Gong API state; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --taskId, --userId
  • tasks list - List user's tasks (/v2/tasks) [intent=direct_read availability=implemented]; approval: none: read-only POST query with schema-gated body; risk: bounded typed POST read-query; response size is capped and secret/download/content-shaped fields are redacted; notes: Executes through the typed operation direct-read engine; no raw body or generic HTTP flag is exposed.; flags: --user-id, --status, --task-action, --task-type, --workspace-id
• Digital Interactions
  • digital-interaction create - sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval [intent=reverse_etl availability=partial write=add_digital_interaction]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: sends call, engagement, or digital interaction content to Gong; requires reverse ETL approval; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --content, --eventId, --eventType, --timestamp
• Logs
  • logs list - Retrieve logs data by type and time range (/v2/logs) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --logType, --fromDateTime, --toDateTime, --cursor
• Coaching
  • coaching list - List all coaching metrics (/v2/coaching) [intent=direct_read availability=implemented]; risk: bounded Gong JSON read; response is limited to 1 MiB and secret/download/content-shaped fields are redacted; flags: --workspace-id, --manager-id, --from, --to
• Outcomes
  • call-outcomes list - List Gong call outcomes as ETL records. [intent=etl availability=implemented stream=call_outcomes]
• Integration Settings
  • integration-settings update - administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation [intent=reverse_etl availability=partial write=integration_settings]; approval: Use reverse ETL plan -> preview -> approval -> execute. Connector command execution is metadata-only for complex object/array records; use typed reverse ETL records.; risk: high: administrative Gong settings or permissions mutation; requires reverse ETL approval and destructive confirmation; notes: No raw HTTP body is accepted. Object and array payloads must come from typed reverse-ETL records validated by writes.json.; flags: --integrationTypeSettings
• Help topics:
  • gong-auth - Use Gong access key and access key secret via credentials; never pass secrets in command text.
  • gong-writes - Gong mutations are typed reverse-ETL actions with plan, preview, approval, execute gates.
  • gong-direct-read - Gong direct reads are bounded JSON GET or typed POST operations with redaction for secret/download/content-shaped fields; call transcripts support a 16 MiB cap.

Commands

Inspect as a manual

pm connectors inspect gong

Inspect as structured JSON

pm connectors inspect gong --json

Agent Rules

• Run pm connectors inspect gong before creating credentials or plans.
• Use --json only when the caller needs structured output; use the manual for human-readable guidance.
• Never ask the user to paste secret values into chat.
• For reverse ETL writes, create a plan, show the preview, wait for explicit approval, then run with the approval token.