"INFINITI" matching MCP tools:

• create_audienceA

  Infinite AudienceA

  Create an audience from one or more existing segments. Audiences are the delivery payload — they have no subtype. To create typed cohorts (filter, matched, similarity, propensity), use create_segment instead. Accepts segment_refs: [{segment_id, role: include|exclude}] to compose the audience with per-segment inclusion/exclusion control. set_logic controls how include-role segments are combined (union or intersection). Returns audience_id, status and campaign_id immediately. Use record_count only for a verified count of this exact composition. Optional campaign_id links the new audience to that campaign (an audience belongs to at most one); omit it to leave the audience unlinked — link it later with link_audience_to_campaign. Requires 'creator' scope.

  ConnectorOAuth

• list_deliveriesA

  Infinite AudienceA

  List up to 100 deliveries for one accessible saved audience. This audience-scoped path checks the audience and its member segments before returning history. Signed download links are omitted; call get_delivery for one completed artifact. Use list_all_deliveries for an org-wide paginated search. Requires 'discovery' scope.

  ConnectorOAuth

• build_saved_integration_segmentA

  Infinite AudienceA

  Create a matched segment, or add unique saved identities to a compatible standalone matched segment. Call preview_saved_integration_segment first and confirm the exact counts and target with the user. Adding to an existing segment replaces its current version. Completion or failure may notify configured organization webhooks, subscriptions, and opted-in user email. This never reruns matching, charges again, or creates an Audience automatically. Requires 'creator' scope.

  Connector

  Destructive

  OAuth

• icx_recall_scopedA

  Calera Infinite Context (ICX) — Persistent Memory & Associative RecallB

  Performs semantic search across persistent long-term memory to retrieve grounded facts, relevant citations, and source documents matching a natural language question. Use this tool when answering questions about previously saved project specifications, codebase decisions, user preferences, or documents. Do NOT use for saving new facts (use icx_remember), do NOT use for raw keyword lookup (use icx_search_facts), and do NOT use for exact character-for-character verbatim quotes (use icx_quote_slot). Read-only operation with zero side effects. Requires X-License-Key authentication.

  ConnectorNo auth

• icx_sync_deltaA

  Calera Infinite Context (ICX) — Persistent Memory & Associative RecallB

  Extracts fine-grained deltas between previous text/document state and updated state, and crystallizes them into the Volumetric Memory lattice in sub-2ms on CPU with automatic secret scrubbing and SHA-256 deduplication. Use this tool for continuous incremental sync of code files, documentation, PRDs, and database schemas. Mutating operation. Requires X-License-Key authentication.

  ConnectorNo auth

• wait_for_deliveryA

  Infinite AudienceA

  Block until a delivery completes or fails. Call this immediately after deliver_audience or deliver_file_match when the response has status: "processing". Polls GET /v1/audiences/{audience_id}/deliveries/{delivery_id} every 15 seconds. On success returns the full delivery object including download_urls[] — share these directly with the user. Each signed link is valid for 24 hours; expires_at is the delivery artifact retention date, not the link expiry. Call get_delivery for a fresh link while the artifact remains available. Throws DELIVERY_FAILED if the export failed (check failure_reason), or DELIVERY_TIMEOUT if it did not finish within timeout_seconds (default 30 seconds). A wait timeout does not cancel or fail the job; use the corresponding get tool to check it again. Requires 'discovery' scope.

  ConnectorOAuth

• wait_for_segmentA

  Infinite AudienceA

  Poll a segment until its status is "active" or "failed". Use after create_segment(subtype=matched) once the file has been uploaded AND confirm_column_mappings has succeeded — the segment will not progress past "pending" before that (the upload is staged). Do not call this before confirm_column_mappings. Also use after create_segment(subtype=similarity/propensity) to wait for the async workflow to complete. Polls GET /v1/segments/{segment_id} every 15 seconds. On success returns the customer-facing segment details — pass segment_id in create_audience.segment_refs, then use the returned audience_id to quote_delivery and deliver_audience. Throws SEGMENT_FAILED if processing failed, or SEGMENT_TIMEOUT if it did not complete within timeout_seconds (default 30 seconds). A wait timeout does not cancel or fail the job; use the corresponding get tool to check it again. Requires 'discovery' scope.

  ConnectorOAuth

• confirm_column_mappingsA

  Infinite AudienceA

  Confirm column mappings for a staged matched-segment upload, starting identity resolution. Call this after analyze_audience_upload has been reviewed with the user and they approve (with or without edits). This moves the staged file into place — nothing runs before this call succeeds. Only call wait_for_segment after this returns ok. Pass exactly one of segment_id or audience_id — use audience_id when only the audience is known (it resolves to that audience's matched member segment server-side). When using audience_id, column_mappings is required (no auto-inference fallback on that path). When using segment_id, column_mappings may be omitted — the platform re-analyzes the active upload itself and accepts its own suggested mapping as-is. Requires 'creator' scope.

  Connector

  Destructive

  OAuth

• cancel_matched_uploadA

  Infinite AudienceA

  Abort a staged matched-segment upload without confirming. Call this when the user declines to proceed after reviewing analyze_audience_upload's results. Clears the staged upload — no file move, no status/mapping-state change. The existing segment is retained; cancellation does not delete it or complete matching. Inspect it with get_segment and archive it with archive_segment if no longer needed. Idempotent. Pass exactly one of segment_id or audience_id — use audience_id when only the audience is known (it resolves to that audience's matched member segment server-side). Requires 'creator' scope.

  Connector

  Destructive

  OAuth

• create_segmentA

  Infinite AudienceA

  Create a segment of the specified subtype. Segments are the typed cohort definitions — audiences are composition wrappers built from segments. filter: reusable filter definition — returns immediately with status active and an unset count unless count_snapshot is supplied. Call count_segment to calculate its saved count. similarity: provide seed_description (natural language) OR seed_segment_id — triggers async icp-segment workflow, returns status pending + workflow_run_id. Call wait_for_segment(segment_id) to block until active. propensity: provide positive_class_segment_id — triggers async ml-model workflow, returns status pending + workflow_run_id. Training can take 10–60 min. Call wait_for_segment(segment_id) to monitor. matched: upload a file — returns upload_url (30-min presigned), or upload_urls (an array, one per shard) when shard_count > 1. The upload is staged (not resolved automatically): PUT the raw file bytes to each returned URL with Content-Type matching file_format: csv=text/csv, avro=application/avro, json=application/json, jsonl=application/x-ndjson. Do not include the MCP bearer token in this upload. After uploading (all shards, if multiple), call analyze_audience_upload(segment_id) (analyzes the first shard only when sharded — all shards must share one schema), present the mappings to the user and wait for their response, then confirm_column_mappings(segment_id) to start resolution (or cancel_matched_upload(segment_id) if they decline) — only THEN call wait_for_segment(segment_id). Upload, analysis, and completed matched resolution emit no usage; the first successful delivery emits one aggregate platform_match usage set together with normal delivery usage. HITL: similarity and propensity default to hitl=false (auto-proceed through all gates). Set hitl=true to pause at review gates. Use file_format to specify csv (default), avro, json, or jsonl and compression=gzip for gzip bytes. Set shard_count > 1 for multiple same-format files; all shards must use the same compression and every CSV shard must include its own header row. Set create_audience=true to automatically wrap the new segment in an audience for immediate delivery. Optional campaign_id records the campaign the segment is built for and links that wrapper audience to it; omit it to leave both outside any campaign. Requires 'creator' scope.

  Connector

  Destructive

  OAuth

• refresh_segmentA

  Infinite AudienceA

  Re-derive a single stale or expired segment from the current IAG. Updates ONLY the segment — referencing audiences are never touched; the response lists them in affected_audiences so you can recount or refresh them (use refresh_audience for a whole-composition refresh). Behaviour is subtype-specific: - filter: synchronous — bumps current_version, nulls record_count, resets the 90-day TTL, returns status: active. If the segment is already fresh the call is a no-op (refreshed: false); pass force: true to bump anyway. Call count_segment afterwards for a fresh count. - matched: with hitl false/unset (default): returns status: pending + upload_url (24 h presign) targeting the canonical upload path — or upload_urls (an array) when shard_count > 1 — instruct the user to re-upload their identity list (one same-format file per URL for a sharded upload; every CSV shard needs its own header); identity resolution runs automatically once all shards are uploaded. Then call wait_for_segment(segment_id). With hitl: true: the upload is staged instead — after upload, follow the analyze_audience_upload -> confirm_column_mappings / cancel_matched_upload -> wait_for_segment sequence documented on create_segment(subtype=matched). Returns 409 SHARED_SOURCE_DATA for matched duplicates that reference another segment's data — the message names the source segment to refresh instead. - similarity: returns status: pending + workflow_run_id (re-launched scoring workflow). Call wait_for_workflow(workflow_run_id). - propensity: returns status: pending + workflow_run_id. 422 MISSING_POSITIVE_CLASS_ID for legacy segments that predate positive_class_segment_id storage — archive those with archive_segment and create a replacement. Returns 409 SEGMENT_ARCHIVED for archived segments — restore them first with archive_segment(restore=true). Requires 'creator' scope.

  Connector

  Destructive

  OAuth

• analyze_audience_uploadA

  Infinite AudienceA

  After uploading a file to the presigned upload_url (or all same-format shards to upload_urls) returned by create_segment(subtype=matched), call this to run LLM-assisted column mapping inference. For a sharded upload, analysis runs against the first shard only — every shard is assumed to share the same schema; a mismatch fails the batch job later with a specific error, not silently here. Returns: segment_id, analysis.columns[] (each: source_column, mapped_to, inferred_type, confidence 0–1, warning), analysis.unmapped_columns[] (column headers that could not be mapped), analysis.recommendations[] (human-readable tips to improve match rates). gzip is detected from file bytes, so a missing or incorrect compression declaration is self-healed during analysis. Returns typed 422 errors for malformed gzip, format mismatch, or malformed input, and 409 FILE_NOT_YET_UPLOADED before upload. The upload is staged (create_segment sets hitl:true) — nothing has started yet. Present these results to the user and wait for their response, then call confirm_column_mappings (to start resolution) or cancel_matched_upload (to abort) — do not call wait_for_segment before confirm_column_mappings succeeds. Requires 'discovery' scope.

  Connector

  Destructive

  OAuth

• list_audiencesA

  Infinite AudienceA

  List a page of non-archived saved audiences for this organization, including status, saved count and campaign_id (null when the audience is in no campaign). Use to find an audience_id before get_audience, recount_audience, or delivery; pass a limit of 1–100. Pass campaign_id to list one campaign's audiences, or unlinked=true for audiences in no campaign (not both). A filtered page can hold fewer than limit audiences — keep following next_cursor until it is null. Requires 'discovery' scope.

  ConnectorOAuth

• list_campaignsA

  Infinite AudienceA

  List a page of campaigns for the organisation. Use when the user asks "what campaigns do I have?" or "show my active campaigns". Returns each campaign brief and status; use get_campaign for linked audience details. Pass the returned next_cursor as cursor to fetch another page. Requires 'discovery' scope.

  ConnectorOAuth

• list_saved_integration_matchesA

  Infinite AudienceA

  List saved integration matches available to build into a segment. Returns aggregate counts without raw identities. Refreshes stored collection counts and may initialize collection metadata; does not create segments or charge for records. Use preview_saved_integration_segment before building. Requires 'discovery' scope.

  ConnectorOAuth

• icx_rememberA

  Calera Infinite Context (ICX) — Persistent Memory & Associative RecallB

  Stores text documents, code snippets, architectural decisions, and factual knowledge into persistent long-term memory partitioned by space_id. Use this tool when saving important facts, project conventions, API keys, user preferences, or documentation for future sessions. Do NOT use for retrieving stored memory (use icx_recall_scoped for semantic QA, icx_search_facts for keyword lookup, or icx_quote_slot for verbatim quotes). Ingestion is additive and non-destructive: creates new memory nodes without overwriting or mutating existing records. Requires X-License-Key authentication.

  ConnectorNo auth

• check_topup_statusA

  Infinite AudienceA

  Read verified state for an existing top-up request without creating a charge. If customer action is required, direct the user to Account → Billing in Infinite Audience; this tool does not return a hosted payment URL. It cannot list top-ups: supply the request_id from the original top-up confirmation, not an invoice/payment ID; if unavailable, ask the account owner to review Account → Billing. Requires 'account-read' scope.

  ConnectorOAuth

• deliver_audienceA

  Infinite AudienceA

  Deliver an audience as a download. LiveRamp and Narrative remain unavailable until their partner pushes can confirm data egress; they cannot create delivery or usage charges. Download has no destination surcharge but still carries base/field delivery usage. Supply template_id OR field_list — not both. Call quote_delivery first: pricing depends on the audience version, selected fields, output settings, and current licensing. Returns delivery_id, status, cost, and (for download) download_urls[] once completed. Requires 'creator' scope.

  Connector

  Destructive

  OAuth

• deliver_file_matchA

  Infinite AudienceA

  Trigger download delivery for a file-based identity match job. Always delivers to download — returns a signed URL to the enriched output file. Use this instead of deliver_audience for matched segments created via create_segment(subtype=matched). The match_id is the segment_id returned by create_segment(subtype=matched) — delivery still routes through an audience wrapper internally. Matched segments do not support LiveRamp or Narrative destinations. Supply template_id OR field_list — not both. Use get_field_catalog() to browse options. Requires 'creator' scope.

  Connector

  Destructive

  OAuth

• quote_deliveryA

  Infinite AudienceA

  Get a usage estimate for delivering an audience before creating billable work. A fresh quote creates a pending quote record and replaces the audience current quote reference; it does not purchase or deliver records. Returns TWO cost figures: cost_estimate is the license-aware expected charge (identities this audience already holds an active 12-month license for elsewhere are excluded from the billable count), and max_cost_estimate is the worst-case ceiling assuming nothing is pre-licensed — always quote max_cost_estimate as the ceiling and cost_estimate as the expected price when license_pricing.status is "live" or "cached"; if license_pricing.status is "unavailable", cost_estimate equals max_cost_estimate exactly and only one number needs mentioning. Key response fields: already_delivered (true = this exact delivery is free), cost_estimate.total_cost / max_cost_estimate.total_cost (total USD), cost_estimate.base_cost / match_cost / field_cost / destination_cost (breakdown; match_cost is the one-time deferred file-match component, never license-discounted), billing_count (records that will be billed), sufficient_balance, shortfall. Call this before deliver_audience so the user can review cost. Requires 'discovery' scope.

  Connector

  Destructive

  OAuth

• list_file_match_jobsA

  Infinite AudienceA

  List async file-based identity resolution jobs for the org. Each job corresponds to a matched segment created via create_segment(subtype=matched). Returns match_id, name, status, matching_status, input_record_count, match_count, match_rate, amount_charged, file_format, compression, segment_id, audience_id, audience_status, created_at, expires_at. amount_charged is null until this run first participates in a successful delivery, then shows its one-time pending match-usage estimate; it excludes delivery usage. audience_status (active/archived/null) reports whether the linked audience is still open-able in Library — job history itself always persists, even after the segment or audience is deleted. Cursor-paginated. Filter by status to find pending/active/failed jobs. Requires 'discovery' scope.

  ConnectorOAuth