@zackbart/connecta 0.28.1 → 0.30.0

package/dist/providers/breeze/skill.generated.js

@@ -0,0 +1,9 @@
+// Generated by scripts/generate-providers.mjs; do not edit.
+export const skill = {
+    "content": ".breezechms.com (Breeze ChMS, now Tithely Church Management): people, tags, events, attendance, forms, volunteers, and giving.\n\nChurch purpose: \n\n## Sensitive data: reduce in code\n\nPeople records carry home addresses, phones, birthdates, and family ties, and\ngiving records reveal who gives and how much — religious affiliation by proxy.\nRead them inside `execute_code` and return only the counts, totals, or the\nfew fields the task needs. Never echo whole profiles or donor lists.\n\n## Ids and profile fields\n\n- Ids are numeric strings. Read them from a list tool; never guess one.\n- Events have two ids: an instance id (`list_events` `id`) for one\n  occurrence, and a series `eventId`. Attendance, volunteers, and\n  check-in take the instance id.\n- Person `details` are keyed by profile `fieldId`, not by name. Call\n  `list_profile_fields` once and map ids to names in code. Multiple-choice\n  values and `list_people` filters use option ids from the same call;\n  join several with `-`.\n- `list_people` has no name search. List without details and match names\n  in code, then `get_person` the hits.\n- Projections rename Breeze's snake_case fields to camelCase and drop photo\n  paths and internal columns; `raw: true` returns Breeze's rows.\n\n## Pagination\n\n`list_people` pages by `cursor`: continue while `page.hasMore`. Breeze\nreports no total, so a full page sets `hasMore` and the next may be empty.\n`list_events` and `list_account_log` do not page: `truncated: true`\nmeans narrow the date range. Other lists return everything.\n\n## Giving\n\n`list_contributions` and `list_funds` call endpoints Breeze removed from\nits public reference in 2023 but still serves. They may change or vanish\nwithout notice; a failure there is not your argument's fault. Amounts are\ndecimal strings; `totalAmount` is an exact sum. Online gifts processed by\nTithe.ly arrive through its sync, batched per deposit, so the latest days\nmay be missing. Read the currency from `get_account_summary`.\n\n## Writes and the hatches\n\n- Every write, `add_person`, `assign_tag`, and `record_check_in`\n  included, follows the routing in the `usage` skill. Fetch it with\n  `skills({ name: \"usage\" })` for guest API and trust rules.\n  Those three only add, and `add_person` never deduplicates.\n- `update_person` field values: text, date (M/D/YYYY), or option id in\n  `response`; email, phone, address, and family_role values in `details`.\n- Every Breeze call is a GET, writes included, so the hatches split by\n  endpoint. `breeze_api_get` admits only reviewed read endpoints and their\n  documented parameters, such as `/tags/list_folders`,\n  `/events/list_event`, `/events/attendance/eligible`, `/giving/view`,\n  and `/pledges/list_campaigns`; it refuses anything else. Everything else\n  — event, family, volunteer, and form writes, giving writes, check-out,\n  undocumented parameters — is `breeze_api_mutate`. Paths are below\n  `/api`; `*_json` values may be passed as objects.\n- `list_account_log` actions include person_created, person_updated,\n  person_deleted, tag_assign, tag_unassign, event_created,\n  attendance_deleted, form_entry_updated, contribution_added,\n  contribution_updated, contribution_deleted, and batch_updated.\n\n## Rate limits\n\nBreeze publishes no limit and sends no rate headers; a third party reports\nabout 20 requests per minute. Keep loops sequential and prefer one list call\nover many single reads.\n",
+    "instructionsHeading": "Church instructions",
+    "fragments": {
+        "guide_0": ".breezechms.com (Breeze ChMS, now Tithely Church Management): people, tags, events, attendance, forms, volunteers, and giving.\n\nChurch purpose: ",
+        "guide_1": "\n\n## Sensitive data: reduce in code\n\nPeople records carry home addresses, phones, birthdates, and family ties, and\ngiving records reveal who gives and how much — religious affiliation by proxy.\nRead them inside `execute_code` and return only the counts, totals, or the\nfew fields the task needs. Never echo whole profiles or donor lists.\n\n## Ids and profile fields\n\n- Ids are numeric strings. Read them from a list tool; never guess one.\n- Events have two ids: an instance id (`list_events` `id`) for one\n  occurrence, and a series `eventId`. Attendance, volunteers, and\n  check-in take the instance id.\n- Person `details` are keyed by profile `fieldId`, not by name. Call\n  `list_profile_fields` once and map ids to names in code. Multiple-choice\n  values and `list_people` filters use option ids from the same call;\n  join several with `-`.\n- `list_people` has no name search. List without details and match names\n  in code, then `get_person` the hits.\n- Projections rename Breeze's snake_case fields to camelCase and drop photo\n  paths and internal columns; `raw: true` returns Breeze's rows.\n\n## Pagination\n\n`list_people` pages by `cursor`: continue while `page.hasMore`. Breeze\nreports no total, so a full page sets `hasMore` and the next may be empty.\n`list_events` and `list_account_log` do not page: `truncated: true`\nmeans narrow the date range. Other lists return everything.\n\n## Giving\n\n`list_contributions` and `list_funds` call endpoints Breeze removed from\nits public reference in 2023 but still serves. They may change or vanish\nwithout notice; a failure there is not your argument's fault. Amounts are\ndecimal strings; `totalAmount` is an exact sum. Online gifts processed by\nTithe.ly arrive through its sync, batched per deposit, so the latest days\nmay be missing. Read the currency from `get_account_summary`.\n\n## Writes and the hatches\n\n- Every write, `add_person`, `assign_tag`, and `record_check_in`\n  included, follows the routing in the `usage` skill. Fetch it with\n  `skills({ name: \"usage\" })` for guest API and trust rules.\n  Those three only add, and `add_person` never deduplicates.\n- `update_person` field values: text, date (M/D/YYYY), or option id in\n  `response`; email, phone, address, and family_role values in `details`.\n- Every Breeze call is a GET, writes included, so the hatches split by\n  endpoint. `breeze_api_get` admits only reviewed read endpoints and their\n  documented parameters, such as `/tags/list_folders`,\n  `/events/list_event`, `/events/attendance/eligible`, `/giving/view`,\n  and `/pledges/list_campaigns`; it refuses anything else. Everything else\n  — event, family, volunteer, and form writes, giving writes, check-out,\n  undocumented parameters — is `breeze_api_mutate`. Paths are below\n  `/api`; `*_json` values may be passed as objects.\n- `list_account_log` actions include person_created, person_updated,\n  person_deleted, tag_assign, tag_unassign, event_created,\n  attendance_deleted, form_entry_updated, contribution_added,\n  contribution_updated, contribution_deleted, and batch_updated.\n\n## Rate limits\n\nBreeze publishes no limit and sends no rate headers; a third party reports\nabout 20 requests per minute. Keep loops sequential and prefer one list call\nover many single reads.\n"
+    }
+};

package/dist/providers/{ccb.d.ts → ccb/index.d.ts}

@@ -1,4 +1,4 @@
-import type { Connector, ConnectorCallAdmissionPolicy } from "../types.js";
+import type { ConnectorCallAdmissionPolicy } from "../../types.js";
 /** CCB's v2 media type. The API and its token endpoint both require it. */
 export declare const CCB_MEDIA_TYPE = "application/vnd.ccbchurch.v2+json";
 /**
@@ -100,4 +100,4 @@ export interface CcbOptions {
     maxResultBytes?: number;
 }
 /** A maintained Church Community Builder (Pushpay ChMS) v2 connection. */
-export declare function ccb(id: string, options: CcbOptions): Connector;
+export declare const ccb: import("../../provider.js").ProviderFactory<CcbOptions>;

package/dist/providers/{ccb.js → ccb/index.js}

@@ -1,3 +1,4 @@
+import { skill } from "./skill.generated.js";
 /**
  * Church Community Builder — sold today as Pushpay ChMS — through its v2 JSON
  * REST API, hand-written over `api()`'s downstream OAuth grant.
@@ -22,16 +23,19 @@
  * pledges, scheduled gifts, and financial settings, and that is all this
  * connector reaches.
  *
- * Drift is handled the Vercel way: `scripts/drift/ccb-endpoints.json` records
+ * Drift is handled the Vercel way: `src/providers/ccb/drift.json` records
  * every endpoint a named tool touches, and
  * `npm run providers:check -- --provider ccb` compares those rows with the
  * published, credential-free OpenAPI document at
  * https://docs.pushpay.io/chms-v2/openapi/chms-api-specs.json.
  */
-import { api } from "../connectors/api.js";
-import { defined } from "../connectors/api-connector.js";
-import { guardedFetch, oauthBearer, retryAfterMs, } from "../connectors/guarded-fetch.js";
-import { ConnectorCallError } from "../errors.js";
+import { api } from "../../connectors/api.js";
+import { defined } from "../../connectors/api-connector.js";
+import { guardedFetch, oauthBearer, retryAfterMs, } from "../../connectors/guarded-fetch.js";
+import { ConnectorCallError } from "../../errors.js";
+import { keys, optionsOf } from "../../config-schema.js";
+import { CALL_ADMISSION } from "../../connectors/option-shapes.js";
+import { asProviderFactory } from "../../provider.js";
 /** CCB's v2 media type. The API and its token endpoint both require it. */
 export const CCB_MEDIA_TYPE = "application/vnd.ccbchurch.v2+json";
 /**
@@ -138,9 +142,7 @@ const CCB_ADMISSION = {
     ],
 };
 function asRecord(value) {
-    return value && typeof value === "object" && !Array.isArray(value)
-        ? value
-        : {};
+    return value && typeof value === "object" && !Array.isArray(value) ? value : {};
 }
 function asArray(value) {
     return Array.isArray(value) ? value : [];
@@ -179,9 +181,7 @@ const NAMED_ENDPOINTS = {
     list_schedules: "GET /scheduling/categories/{id}/schedules",
     list_individual_assignments: "GET /individuals/{id}/assignments",
     list_individual_notes: "GET /individuals/{id}/notes",
-    get_giving_metrics: (args) => args["familyId"] !== undefined
-        ? "GET /families/{id}/metrics/giving"
-        : "GET /individuals/{id}/metrics/giving",
+    get_giving_metrics: (args) => args["familyId"] !== undefined ? "GET /families/{id}/metrics/giving" : "GET /individuals/{id}/metrics/giving",
     list_individual_pledges: "GET /individuals/{id}/pledges",
     ccb_api_search: (args) => `POST /search/${typeof args["domain"] === "string" ? args["domain"] : "?"}/results`,
 };
@@ -208,17 +208,16 @@ function endpointPartition(toolName, raw) {
     else if (named)
         key = named(args);
     else {
-        const method = toolName === "ccb_api_get"
-            ? "GET"
-            : typeof args["method"] === "string"
-                ? args["method"]
-                : "?";
+        const method = toolName === "ccb_api_get" ? "GET" : typeof args["method"] === "string" ? args["method"] : "?";
         const path = typeof args["path"] === "string" ? args["path"] : "";
         key = `${method} ${pathTemplate(path)}`;
     }
     // ASCII-only after templating in every realistic case; the slice keeps an
     // absurd path inside the admission controller's 128-byte key limit anyway.
-    return [...key].filter((c) => c.charCodeAt(0) < 128).join("").slice(0, MAX_PARTITION_KEY_CHARS);
+    return [...key]
+        .filter((c) => c.charCodeAt(0) < 128)
+        .join("")
+        .slice(0, MAX_PARTITION_KEY_CHARS);
 }
 // ---------------------------------------------------------------------------
 // Transport and typed failures
@@ -278,7 +277,7 @@ function ccbFailure(status, headers, payload) {
         return new ConnectorCallError("auth_required", `${detail} CCB rejected the connection's access token. Call authorize_connector for this connector, then retry.`);
     }
     if (status === 403) {
-        return new ConnectorCallError("connector_call_failed", `${detail} The grant lacks this endpoint's scope, or (Identity Auth) the signed-in person lacks the permission in CCB. Re-authorizing alone will not fix it.`, { retryable: false });
+        return new ConnectorCallError("provider_permission_denied", `${detail} The grant lacks this endpoint's scope, or (Identity Auth) the signed-in person lacks the permission in CCB. Ask the CCB administrator to grant this endpoint's scope or the signed-in person's permission. Re-authorizing alone will not fix it.`, { retryable: false });
     }
     if (status === 404) {
         return new ConnectorCallError("not_found", `${detail} Re-read the id from its list tool; CCB reports a permission gap as 403, so this record does not exist.`);
@@ -310,7 +309,9 @@ async function callCcb(send, request, ctx) {
             throw ccbFailure(response.status, response.headers, "value" in parsed ? parsed.value : undefined);
         }
         if (!("value" in parsed)) {
-            throw new ConnectorCallError("connector_call_failed", "CCB returned a successful response that is not JSON.", { retryable: false });
+            throw new ConnectorCallError("connector_call_failed", "CCB returned a successful response that is not JSON.", {
+                retryable: false,
+            });
         }
         return { body: parsed.value, headers: response.headers };
     });
@@ -569,9 +570,7 @@ function projectProcess(value) {
         ownerId: row["owner_id"],
         hidden: row["hidden"],
         archived: row["archived"],
-        managers: Array.isArray(row["process_managers"])
-            ? row["process_managers"].map(ref).filter(Boolean)
-            : undefined,
+        managers: Array.isArray(row["process_managers"]) ? row["process_managers"].map(ref).filter(Boolean) : undefined,
     });
 }
 function projectQueue(value) {
@@ -732,88 +731,220 @@ function input(properties, required = []) {
 }
 const PHONES = obj({ mobile: STR, home: STR, work: STR });
 const PERSON_SCHEMA = obj({
-    id: INT, name: STR, firstName: STR, lastName: STR, email: STR, phones: PHONES,
-    birthday: STR, familyId: INT, familyPosition: STR, campusId: INT,
-    membershipTypeId: INT, active: BOOL, deceased: STR,
+    id: INT,
+    name: STR,
+    firstName: STR,
+    lastName: STR,
+    email: STR,
+    phones: PHONES,
+    birthday: STR,
+    familyId: INT,
+    familyPosition: STR,
+    campusId: INT,
+    membershipTypeId: INT,
+    active: BOOL,
+    deceased: STR,
 });
 const ADDRESS = obj({ street: STR, city: STR, state: STR, zip: STR, country: STR });
 const INDIVIDUAL_SCHEMA = obj({
     ...PERSON_SCHEMA.properties,
-    gender: STR, maritalStatus: STR, homeAddress: ADDRESS, mailingAddress: ADDRESS,
-    campusName: STR, membershipTypeName: STR, membershipDate: STR, baptized: BOOL,
-    lastAttendedDate: STR, listed: BOOL, customFields: { type: "object" },
+    gender: STR,
+    maritalStatus: STR,
+    homeAddress: ADDRESS,
+    mailingAddress: ADDRESS,
+    campusName: STR,
+    membershipTypeName: STR,
+    membershipDate: STR,
+    baptized: BOOL,
+    lastAttendedDate: STR,
+    listed: BOOL,
+    customFields: { type: "object" },
+});
+const FAMILY_SCHEMA = obj({
+    id: INT,
+    lastName: STR,
+    address: ADDRESS,
+    members: { type: "array", items: PERSON_SCHEMA },
 });
-const FAMILY_SCHEMA = obj({ id: INT, lastName: STR, address: ADDRESS, members: { type: "array", items: PERSON_SCHEMA } });
 const REF = obj({ id: INT, name: STR });
 const GROUP_SCHEMA = obj({
-    id: INT, name: STR, campusId: INT, inactive: BOOL, groupType: STR,
-    interactionType: STR, membershipType: STR, mainLeader: REF, departmentId: INT,
+    id: INT,
+    name: STR,
+    campusId: INT,
+    inactive: BOOL,
+    groupType: STR,
+    interactionType: STR,
+    membershipType: STR,
+    mainLeader: REF,
+    departmentId: INT,
 });
 const GROUP_DETAIL_SCHEMA = obj({
     ...GROUP_SCHEMA.properties,
-    description: STR, memberCount: INT, childcare: BOOL, listed: BOOL, full: BOOL,
-    meetDay: STR, meetTime: STR, area: STR, address: ADDRESS,
+    description: STR,
+    memberCount: INT,
+    childcare: BOOL,
+    listed: BOOL,
+    full: BOOL,
+    meetDay: STR,
+    meetTime: STR,
+    area: STR,
+    address: ADDRESS,
 });
 const MEMBERSHIP_SCHEMA = obj({
-    groupId: INT, individualId: INT, status: STR, dateAdded: STR, name: STR, email: STR,
-    phones: PHONES, groupName: STR, groupType: STR, groupInactive: BOOL,
+    groupId: INT,
+    individualId: INT,
+    status: STR,
+    dateAdded: STR,
+    name: STR,
+    email: STR,
+    phones: PHONES,
+    groupName: STR,
+    groupType: STR,
+    groupInactive: BOOL,
 });
 const OCCURRENCE_SCHEMA = obj({
-    eventId: INT, occurrence: STR, start: STR, end: STR, name: STR, campusId: INT,
-    recurs: BOOL, groupId: INT, groupName: STR,
+    eventId: INT,
+    occurrence: STR,
+    start: STR,
+    end: STR,
+    name: STR,
+    campusId: INT,
+    recurs: BOOL,
+    groupId: INT,
+    groupName: STR,
 });
 const EVENT_SCHEMA = obj({
-    id: INT, name: STR, description: STR, campusId: INT, start: STR, end: STR,
-    recurs: BOOL, recurEndDate: STR, approvalStatus: STR, address: ADDRESS,
-    guestCounts: { type: "object" }, group: REF,
+    id: INT,
+    name: STR,
+    description: STR,
+    campusId: INT,
+    start: STR,
+    end: STR,
+    recurs: BOOL,
+    recurEndDate: STR,
+    approvalStatus: STR,
+    address: ADDRESS,
+    guestCounts: { type: "object" },
+    group: REF,
     attendance: obj({ status: STR, totalAttendance: INT, visitors: INT }),
 });
 const ATTENDANCE_SCHEMA = obj({
-    eventId: INT, occurrence: STR, status: STR, totalAttendance: INT, visitors: INT, topic: STR,
+    eventId: INT,
+    occurrence: STR,
+    status: STR,
+    totalAttendance: INT,
+    visitors: INT,
+    topic: STR,
 });
 const ATTENDEE_SCHEMA = obj({ individualId: INT, name: STR, familyId: INT, occurrenceDate: STR });
 const FORM_SCHEMA = obj({
-    id: INT, title: STR, status: STR, campusId: INT, start: STR, end: STR, public: BOOL, url: STR,
+    id: INT,
+    title: STR,
+    status: STR,
+    campusId: INT,
+    start: STR,
+    end: STR,
+    public: BOOL,
+    url: STR,
 });
 const RESPONSE_SCHEMA = obj({
-    id: INT, formId: INT, individualId: INT, created: STR, paymentStatus: STR,
+    id: INT,
+    formId: INT,
+    individualId: INT,
+    created: STR,
+    paymentStatus: STR,
     answers: { type: "array", items: obj({ questionId: INT, answer: STR }) },
 });
 const PROCESS_SCHEMA = obj({
-    id: INT, name: STR, campusId: INT, ownerId: INT, hidden: BOOL, archived: BOOL,
+    id: INT,
+    name: STR,
+    campusId: INT,
+    ownerId: INT,
+    hidden: BOOL,
+    archived: BOOL,
     managers: { type: "array", items: REF },
 });
 const QUEUE_SCHEMA = obj({
-    id: INT, name: STR, processId: INT, processName: STR, description: STR,
-    assignedCount: INT, unassignedCount: INT,
+    id: INT,
+    name: STR,
+    processId: INT,
+    processName: STR,
+    description: STR,
+    assignedCount: INT,
+    unassignedCount: INT,
 });
 const QUEUE_INDIVIDUAL_SCHEMA = obj({
-    queueIndividualId: INT, queueId: INT, queueName: STR, processId: INT, processName: STR,
-    individualId: INT, name: STR, status: STR, due: STR, created: STR, completed: STR,
+    queueIndividualId: INT,
+    queueId: INT,
+    queueName: STR,
+    processId: INT,
+    processName: STR,
+    individualId: INT,
+    name: STR,
+    status: STR,
+    due: STR,
+    created: STR,
+    completed: STR,
 });
 const CATEGORY_SCHEMA = obj({
-    id: INT, name: STR, campus: REF, organizer: REF, archived: BOOL,
+    id: INT,
+    name: STR,
+    campus: REF,
+    organizer: REF,
+    archived: BOOL,
     teams: { type: "array", items: REF },
 });
 const SCHEDULE_SCHEMA = obj({
-    id: INT, categoryId: INT, name: STR, start: STR, end: STR, needed: INT,
+    id: INT,
+    categoryId: INT,
+    name: STR,
+    start: STR,
+    end: STR,
+    needed: INT,
     events: { type: "array", items: obj({ id: INT, name: STR, start: STR, end: STR }) },
     metrics: { type: "object" },
 });
 const ASSIGNMENT_SCHEMA = obj({
-    id: INT, categoryId: INT, scheduleId: INT, eventId: INT, eventPositionId: INT,
-    status: STR, statusReason: STR, dateNotified: STR, individualId: INT, name: STR,
+    id: INT,
+    categoryId: INT,
+    scheduleId: INT,
+    eventId: INT,
+    eventPositionId: INT,
+    status: STR,
+    statusReason: STR,
+    dateNotified: STR,
+    individualId: INT,
+    name: STR,
 });
 const NOTE_SCHEMA = obj({
-    id: INT, individualId: INT, note: STR, date: STR, sharingLevel: STR,
-    context: STR, contextId: INT, creator: REF,
+    id: INT,
+    individualId: INT,
+    note: STR,
+    date: STR,
+    sharingLevel: STR,
+    context: STR,
+    contextId: INT,
+    creator: REF,
 });
 const PLEDGE_SCHEMA = obj({
-    id: INT, campusId: INT, fund: STR, pledged: { type: "number" }, paid: { type: "number" },
-    expected: { type: "number" }, total: { type: "number" }, frequency: STR, start: STR, end: STR,
+    id: INT,
+    campusId: INT,
+    fund: STR,
+    pledged: { type: "number" },
+    paid: { type: "number" },
+    expected: { type: "number" },
+    total: { type: "number" },
+    frequency: STR,
+    start: STR,
+    end: STR,
 });
 const ME_SCHEMA = obj({
-    id: INT, name: STR, email: STR, username: STR, familyId: INT, campusId: INT,
+    id: INT,
+    name: STR,
+    email: STR,
+    username: STR,
+    familyId: INT,
+    campusId: INT,
     userTypes: { type: "array", items: STR },
 });
 /** Domains with a `POST /search/{domain}/results` endpoint in the v2 spec. */
@@ -952,10 +1083,19 @@ function tools(send, defaultPerPage, writable) {
             description: "Find CCB individuals by partial name, phone, or email. Returns lean profiles with family and campus ids; not the advanced search.",
             annotations: readOnly,
             inputSchema: input({
-                query: { type: "string", minLength: 1, maxLength: 200, description: "Partial name, phone number, or email. Omit to page everyone." },
+                query: {
+                    type: "string",
+                    minLength: 1,
+                    maxLength: 200,
+                    description: "Partial name, phone number, or email. Omit to page everyone.",
+                },
                 includeInactive: { type: "boolean", description: "Include inactive individuals." },
                 includeDeceased: { type: "boolean", description: "Include deceased individuals." },
-                campusScope: { type: "string", enum: ["all", "current", "other"], description: "Campuses to search relative to the connection's own." },
+                campusScope: {
+                    type: "string",
+                    enum: ["all", "current", "other"],
+                    description: "Campuses to search relative to the connection's own.",
+                },
                 page: PAGE,
                 perPage,
                 raw: RAW,
@@ -972,7 +1112,9 @@ function tools(send, defaultPerPage, writable) {
             name: "get_individual",
             description: "Get one CCB individual's profile: contact, addresses, membership, and custom fields. CCB allows one call a second; prefer list_individuals for many people.",
             annotations: readOnly,
-            inputSchema: input({ individualId: idProperty("Individual id from list_individuals."), raw: RAW }, ["individualId"]),
+            inputSchema: input({ individualId: idProperty("Individual id from list_individuals."), raw: RAW }, [
+                "individualId",
+            ]),
             outputSchema: INDIVIDUAL_SCHEMA,
             handler: (args, ctx) => one(`/individuals/${id(args["individualId"])}`, args, ctx, projectIndividual),
         },
@@ -1076,7 +1218,11 @@ function tools(send, defaultPerPage, writable) {
             annotations: readOnly,
             inputSchema: input({
                 name: { type: "string", minLength: 1, maxLength: 200, description: "Partial event name." },
-                range: { type: "string", enum: ["ALL", "PAST", "CURRENT", "FUTURE"], description: "Time range relative to now." },
+                range: {
+                    type: "string",
+                    enum: ["ALL", "PAST", "CURRENT", "FUTURE"],
+                    description: "Time range relative to now.",
+                },
                 page: PAGE,
                 perPage,
                 raw: RAW,
@@ -1107,7 +1253,10 @@ function tools(send, defaultPerPage, writable) {
             name: "get_event_attendance",
             description: "Get one CCB event occurrence's attendance summary: met or not, totals, visitors, topic. Drops leader notes and prayer requests unless raw.",
             annotations: readOnly,
-            inputSchema: input({ eventId: idProperty("Event id from list_events."), occurrence: OCCURRENCE, raw: RAW }, ["eventId", "occurrence"]),
+            inputSchema: input({ eventId: idProperty("Event id from list_events."), occurrence: OCCURRENCE, raw: RAW }, [
+                "eventId",
+                "occurrence",
+            ]),
             outputSchema: ATTENDANCE_SCHEMA,
             handler: (args, ctx) => one(`/events/${id(args["eventId"])}/attendance/${compactDate(args["occurrence"])}`, args, ctx, projectAttendance),
         },
@@ -1125,7 +1274,11 @@ function tools(send, defaultPerPage, writable) {
             annotations: readOnly,
             inputSchema: input({
                 title: { type: "string", minLength: 1, maxLength: 200, description: "Partial form title." },
-                status: { type: "string", enum: ["ACTIVE", "SCHEDULED", "UNPUBLISHED", "EXPIRED", "ARCHIVED"], description: "Form status." },
+                status: {
+                    type: "string",
+                    enum: ["ACTIVE", "SCHEDULED", "UNPUBLISHED", "EXPIRED", "ARCHIVED"],
+                    description: "Form status.",
+                },
                 campusId: idProperty("Only forms at this campus."),
                 page: PAGE,
                 perPage,
@@ -1187,7 +1340,11 @@ function tools(send, defaultPerPage, writable) {
             description: "List people in CCB process queues the connection can see, by status or assignment. CCB offers no per-queue filter here; filter on queueId.",
             annotations: readOnly,
             inputSchema: input({
-                status: { type: "string", enum: ["NOT_STARTED", "WAITING", "IN_PROCESS", "DONE"], description: "Queue status." },
+                status: {
+                    type: "string",
+                    enum: ["NOT_STARTED", "WAITING", "IN_PROCESS", "DONE"],
+                    description: "Queue status.",
+                },
                 assignment: { type: "string", enum: ["ASSIGNED_TO_ME", "UNASSIGNED"], description: "Omit for both." },
                 page: PAGE,
                 perPage,
@@ -1205,12 +1362,20 @@ function tools(send, defaultPerPage, writable) {
             annotations: readOnly,
             inputSchema: input({
                 search: { type: "string", minLength: 1, maxLength: 200, description: "Category name search." },
-                status: { type: "string", enum: ["ACTIVE", "ARCHIVED", "ACTIVE_AND_ARCHIVED"], description: "Defaults to CCB's own (active)." },
+                status: {
+                    type: "string",
+                    enum: ["ACTIVE", "ARCHIVED", "ACTIVE_AND_ARCHIVED"],
+                    description: "Defaults to CCB's own (active).",
+                },
                 raw: RAW,
             }),
             outputSchema: rowsOutput("categories", CATEGORY_SCHEMA),
             handler: async (args, ctx) => {
-                const { body } = await callCcb(send, { method: "GET", path: "/scheduling/categories", query: { search_term: args["search"], status: args["status"] } }, ctx);
+                const { body } = await callCcb(send, {
+                    method: "GET",
+                    path: "/scheduling/categories",
+                    query: { search_term: args["search"], status: args["status"] },
+                }, ctx);
                 const rows = Array.isArray(body) ? body : body && typeof body === "object" ? [body] : [];
                 return { categories: args["raw"] === true ? rows : rows.map(projectCategory) };
             },
@@ -1261,7 +1426,11 @@ function tools(send, defaultPerPage, writable) {
             annotations: readOnly,
             inputSchema: input({
                 individualId: idProperty("Individual id."),
-                context: { type: "string", enum: ["GROUP", "DEPARTMENT", "PROCESS_QUEUE", "GENERAL"], description: "Only notes in this context." },
+                context: {
+                    type: "string",
+                    enum: ["GROUP", "DEPARTMENT", "PROCESS_QUEUE", "GENERAL"],
+                    description: "Only notes in this context.",
+                },
                 page: PAGE,
                 perPage,
                 raw: RAW,
@@ -1286,7 +1455,11 @@ function tools(send, defaultPerPage, writable) {
             outputSchema: obj({
                 periods: {
                     type: "array",
-                    items: obj({ start: STR, count: INT, members: { type: "array", items: obj({ individualId: INT, count: INT }) } }),
+                    items: obj({
+                        start: STR,
+                        count: INT,
+                        members: { type: "array", items: obj({ individualId: INT, count: INT }) },
+                    }),
                 },
             }),
             handler: async (args, ctx) => {
@@ -1327,7 +1500,9 @@ function tools(send, defaultPerPage, writable) {
             name: "list_individual_pledges",
             description: "List one CCB individual's pledges: fund, pledged, paid, and expected amounts, and recurrence.",
             annotations: readOnly,
-            inputSchema: input({ individualId: idProperty("Individual id."), page: PAGE, perPage, raw: RAW }, ["individualId"]),
+            inputSchema: input({ individualId: idProperty("Individual id."), page: PAGE, perPage, raw: RAW }, [
+                "individualId",
+            ]),
             outputSchema: listOutput("pledges", PLEDGE_SCHEMA),
             handler: (args, ctx) => list(`/individuals/${id(args["individualId"])}/pledges`, args, ctx, "pledges", projectPledge),
         },
@@ -1392,7 +1567,11 @@ function tools(send, defaultPerPage, writable) {
             description: "Call any CCB v2 JSON POST, PUT, PATCH, or DELETE endpoint. The approval-gated hatch for every write; no file uploads.",
             annotations: { readOnlyHint: false, destructiveHint: true },
             inputSchema: input({
-                method: { type: "string", enum: ["POST", "PUT", "PATCH", "DELETE"], description: "HTTP method the endpoint documents." },
+                method: {
+                    type: "string",
+                    enum: ["POST", "PUT", "PATCH", "DELETE"],
+                    description: "HTTP method the endpoint documents.",
+                },
                 path: HATCH_PATH,
                 query: QUERY,
                 body: {
@@ -1430,52 +1609,9 @@ function usageGuide(purpose, environment, mode, scopes, writable, instructions)
     const extra = instructions?.trim();
     return `# Church Community Builder (Pushpay ChMS)
 
-${routingLine(environment, mode, writable)}: ${purpose}
-
-## Resolve ids before acting
-
-- Every record is addressed by an integer id. Read it from a list tool and
-  pass it back; never guess one. \`list_individuals\` finds people by partial
-  name, phone, or email; \`familyId\`, \`groupId\`, \`eventId\`, form, process,
-  queue, and category ids come from their own list tools.
-- Events are series with dated occurrences. \`list_events\` returns
-  \`eventId\` plus \`occurrence\`; the attendance tools need both.
-- Campus and membership-type ids resolve through \`ccb_api_get\` on
-  \`/campuses\` and \`/church/membership_types\`.
-${mode === "identity" ? "- Results are what the signed-in person may see in CCB. A 403 means that person lacks the permission, not that the connection is broken.\n" : ""}
-## Sensitive data stays out of results
-
-Projections drop allergies, giving numbers and dates, attendance prayer
-requests and leader notes, and form payment details; \`raw: true\` returns
-CCB's full row. Notes, pledges, giving metrics, background checks, and
-financial settings are pastoral or financial: read them inside
-\`execute_code\` and return only the reduced answer the task needs.
-
-Giving in v2 is counts per period (\`get_giving_metrics\`), pledges, scheduled
-gifts, and financial settings. Transactions, batches, and amounts per gift
-live only in CCB's legacy v1 XML API, which this connector does not reach.
-
-## Paging
-
-Lists take \`page\` (1-based) and \`perPage\` (25, 50, 75, or 100 only) and
-return \`page.hasMore\`, \`page.nextPage\`, and \`page.total\`. Field names are
-CCB's snake_case turned camelCase.
-
-## Rate limits
-
-CCB allows each endpoint about one call a second after a burst of 60, per API
-client. This connector keeps a per-runtime 60-per-minute window per endpoint
-— an approximation, not an enforcement. \`get_individual\` (1/s),
-\`list_schedules\` (1 per 2 s), and individuals advanced search (1 per 5 s)
-allow no burst: page a list instead of looping a get, and space those calls.
-A \`rate_limited\` failure carries CCB's wait.
-
-## Raw access
-
-\`ccb_api_get\` reaches any GET path; \`ccb_api_search\` runs advanced searches,
-which CCB serves as POST but are reads.${writable ? " `ccb_api_mutate` reaches every POST, PUT, PATCH, and DELETE and always needs approval." : " This connection is read-only, so it ships no write tool."} Paths are below the API origin, such as \`/church/membership_types\`.
+${routingLine(environment, mode, writable)}: ${purpose}${skill.fragments.guide_0}${mode === "identity" ? "- Results are what the signed-in person may see in CCB. A 403 means that person lacks the permission, not that the connection is broken.\n" : ""}${skill.fragments.guide_1}${writable ? " `ccb_api_mutate` reaches every POST, PUT, PATCH, and DELETE and always needs approval." : " This connection is read-only, so it ships no write tool."} Paths are below the API origin, such as \`/church/membership_types\`.
 Granted scopes: ${scopes.join(" ")}. A call outside them fails with 403.
-${extra ? `\n## Church instructions\n\n${extra}\n` : ""}`;
+${extra ? `\n## ${skill.instructionsHeading}\n\n${extra}\n` : ""}`;
 }
 // ---------------------------------------------------------------------------
 // Constructor
@@ -1513,8 +1649,27 @@ function defaultTitle(environment, mode, writable) {
     ];
     return `Pushpay ChMS (CCB)${qualifiers.length ? ` — ${qualifiers.join(", ")}` : ""}`;
 }
+/** The closed options ccb() accepts; see `assertKnownOptions`. */
+const CCB_OPTIONS = optionsOf()({
+    ...keys("purpose", "title", "instructions", "environment", "mode", "clientId", "clientSecret", "access", "scopes", "subdomain", "defaultPerPage", "maxResultBytes"),
+    callAdmission: CALL_ADMISSION,
+});
 /** A maintained Church Community Builder (Pushpay ChMS) v2 connection. */
-export function ccb(id, options) {
+export const ccb = asProviderFactory({
+    name: "ccb",
+    title: "Church Community Builder",
+    kind: "api",
+    readme: "Church Community Builder",
+    bundle: {
+        "baselineGzip": 120479,
+        "maxGzip": 180479,
+        "note": "./providers/ccb starts at 120,479 B gzip: the hand-written surface plus api()'s static OAuth grant over the SDK's auth client, with no remoteMcp() transport. The cap uses the baseline + 60,000 B policy.",
+    },
+    skill,
+    options: CCB_OPTIONS,
+    create: ccbConnector,
+});
+function ccbConnector(id, options) {
     const purpose = typeof options.purpose === "string" ? options.purpose.trim() : "";
     if (!purpose)
         throw new Error("ccb() requires a non-empty church purpose.");

package/dist/providers/ccb/skill.generated.js

@@ -0,0 +1,9 @@
+// Generated by scripts/generate-providers.mjs; do not edit.
+export const skill = {
+    "content": "\n\n## Resolve ids before acting\n\n- Every record is addressed by an integer id. Read it from a list tool and\n  pass it back; never guess one. `list_individuals` finds people by partial\n  name, phone, or email; `familyId`, `groupId`, `eventId`, form, process,\n  queue, and category ids come from their own list tools.\n- Events are series with dated occurrences. `list_events` returns\n  `eventId` plus `occurrence`; the attendance tools need both.\n- Campus and membership-type ids resolve through `ccb_api_get` on\n  `/campuses` and `/church/membership_types`.\n\n## Sensitive data stays out of results\n\nProjections drop allergies, giving numbers and dates, attendance prayer\nrequests and leader notes, and form payment details; `raw: true` returns\nCCB's full row. Notes, pledges, giving metrics, background checks, and\nfinancial settings are pastoral or financial: read them inside\n`execute_code` and return only the reduced answer the task needs.\n\nGiving in v2 is counts per period (`get_giving_metrics`), pledges, scheduled\ngifts, and financial settings. Transactions, batches, and amounts per gift\nlive only in CCB's legacy v1 XML API, which this connector does not reach.\n\n## Paging\n\nLists take `page` (1-based) and `perPage` (25, 50, 75, or 100 only) and\nreturn `page.hasMore`, `page.nextPage`, and `page.total`. Field names are\nCCB's snake_case turned camelCase.\n\n## Rate limits\n\nCCB allows each endpoint about one call a second after a burst of 60, per API\nclient. This connector keeps a per-runtime 60-per-minute window per endpoint\n— an approximation, not an enforcement. `get_individual` (1/s),\n`list_schedules` (1 per 2 s), and individuals advanced search (1 per 5 s)\nallow no burst: page a list instead of looping a get, and space those calls.\nA `rate_limited` failure carries CCB's wait.\n\n## Raw access\n\n`ccb_api_get` reaches any GET path; `ccb_api_search` runs advanced searches,\nwhich CCB serves as POST but are reads.",
+    "instructionsHeading": "Church instructions",
+    "fragments": {
+        "guide_0": "\n\n## Resolve ids before acting\n\n- Every record is addressed by an integer id. Read it from a list tool and\n  pass it back; never guess one. `list_individuals` finds people by partial\n  name, phone, or email; `familyId`, `groupId`, `eventId`, form, process,\n  queue, and category ids come from their own list tools.\n- Events are series with dated occurrences. `list_events` returns\n  `eventId` plus `occurrence`; the attendance tools need both.\n- Campus and membership-type ids resolve through `ccb_api_get` on\n  `/campuses` and `/church/membership_types`.\n",
+        "guide_1": "\n## Sensitive data stays out of results\n\nProjections drop allergies, giving numbers and dates, attendance prayer\nrequests and leader notes, and form payment details; `raw: true` returns\nCCB's full row. Notes, pledges, giving metrics, background checks, and\nfinancial settings are pastoral or financial: read them inside\n`execute_code` and return only the reduced answer the task needs.\n\nGiving in v2 is counts per period (`get_giving_metrics`), pledges, scheduled\ngifts, and financial settings. Transactions, batches, and amounts per gift\nlive only in CCB's legacy v1 XML API, which this connector does not reach.\n\n## Paging\n\nLists take `page` (1-based) and `perPage` (25, 50, 75, or 100 only) and\nreturn `page.hasMore`, `page.nextPage`, and `page.total`. Field names are\nCCB's snake_case turned camelCase.\n\n## Rate limits\n\nCCB allows each endpoint about one call a second after a burst of 60, per API\nclient. This connector keeps a per-runtime 60-per-minute window per endpoint\n— an approximation, not an enforcement. `get_individual` (1/s),\n`list_schedules` (1 per 2 s), and individuals advanced search (1 per 5 s)\nallow no burst: page a list instead of looping a get, and space those calls.\nA `rate_limited` failure carries CCB's wait.\n\n## Raw access\n\n`ccb_api_get` reaches any GET path; `ccb_api_search` runs advanced searches,\nwhich CCB serves as POST but are reads."
+    }
+};