@noresponse/mcp 0.13.24 → 0.13.25

package/CHANGELOG.md

@@ -1,5 +1,9 @@
 # Changelog
 
+## 0.13.25
+
+No changes detected.
+
 ## 0.13.24
 
 * `SDK.settings.get()`:  `response.refused_sources[].enum(life360_phone)` **Added** (Breaking ⚠️)

package/README.md

@@ -2,14 +2,21 @@
 
 # NoResponse MCP server
 
-An [MCP](https://modelcontextprotocol.io) server for the NoResponse API, so an agent can assess an account creation, login or purchase, wait for a pending assessment, page through an account's history and record what you later learned. It is generated from the same OpenAPI document as the NoResponse SDKs.
+An [MCP](https://modelcontextprotocol.io) server for the NoResponse API, so an agent can investigate people: create a group, add people, screen them, read every account and leaked row that the lookups found, and search again on a value. It also assesses an account creation, login or purchase and records what you later learned. It is generated from the same OpenAPI document as the NoResponse SDKs.
 
-The server uses the API key you give it and stores nothing. API key rotation and key creation are not tools: rotate and create keys from your own backend. The integration tools need a key with the `integrations` scope, the key, settings and list tools a key with the `management` scope, and the event tools a key with the `events` scope; give the server a key without them to keep them unusable.
+The server uses the API key you give it and stores nothing. API key rotation and key creation are not tools: rotate and create keys from your own backend. The integration tools need a key with the `integrations` scope, the key, settings and list tools a key with the `management` scope, and the group, people, search and screening tools a key with the `groups` scope; give the server a key without them to keep them unusable.
 
 ## Tools
 
 | Tool | What it does | API call |
 |---|---|---|
+| `create_group` | Create a group: one list of people to investigate or screen together. Needs the `groups` scope. | `POST /v1/workspaces/{workspace_id}/groups/{environment}` |
+| `add_people` | Add or update up to 500 people, each with a name and the identifiers you have. Needs the `groups` scope. | `POST .../groups/{environment}/{group_id}/people` |
+| `start_screening` | Run the lookups and the analysis for the group's people. Needs the `groups` scope. | `POST .../groups/{environment}/{group_id}/screenings` |
+| `get_screening` | Read a screening's status and progress. Needs the `groups` scope. | `GET .../screenings/{screening_id}` |
+| `get_person` | Read a person's report: every account and leaked row as a record with its source, and the findings. Needs the `groups` scope. | `GET .../people/{person_id}` |
+| `start_search` | Search again on a value of one of the person's records. Needs the `groups` scope. | `POST .../people/{person_id}/searches` |
+| `review_person` | Record your team's decision on a person: `cleared`, `watch` or `escalated`. Needs the `groups` scope. | `POST .../people/{person_id}/review` |
 | `create_assessment` | Assess identifiers (email, phone, username, name, IP) for an `account_creation`, `login` or `purchase` flow. Sends a fresh Idempotency-Key when you give none. | `POST /v1/assessments` |
 | `get_assessment` | Read the latest version of one assessment. | `GET /v1/assessments/{assessment_id}` |
 | `wait_for_assessment` | Poll a pending assessment, at the interval Retry-After asks for, until it settles or `timeout_seconds` (default 25, at most 55) pass. | `GET /v1/assessments/{assessment_id}` |
@@ -41,7 +48,7 @@ The server uses the API key you give it and stores nothing. API key rotation and
 | `create_list_entry` | Add an email or email domain to the block or allow list; the accounts it matches are decided again. Needs the `management` scope. | `POST /v1/workspaces/{workspace_id}/lists/{environment}` |
 | `delete_list_entry` | Delete an entry; the accounts it matched go back to their scored decision. Needs the `management` scope. | `DELETE /v1/workspaces/{workspace_id}/lists/{environment}/{list_entry_id}` |
 
-The table shows the most used tools. [Connect an agent](https://docs.noresponse.org/guides/agents) lists every tool with its required scope, including the review, evaluation and event tools.
+The table shows the most used tools. [Connect an agent](https://docs.noresponse.org/guides/agents) lists every tool with its required scope, including the other group, review and evaluation tools.
 
 Act on `account.risk.decision` (`allow`, `step_up`, `review`, `block`, or `held` while a held flow waits for its investigation), the account's current decision for the assessment's flow. `detail` holds the assessment's own risk, recommendation and `assessment_version`. While the account's first assessment is pending, `account` is absent, so apply your own fallback for that, 429, 503 or a timeout.
 

package/bin/mcp-server.js

@@ -48136,9 +48136,9 @@ var init_config = __esm(() => {
   SDK_METADATA = {
     language: "typescript",
     openapiDocVersion: "1",
-    sdkVersion: "0.13.24",
+    sdkVersion: "0.13.25",
     genVersion: "internal",
-    userAgent: "noresponse-sdk/mcp-typescript 0.13.24 internal 1 @noresponse/mcp"
+    userAgent: "noresponse-sdk/mcp-typescript 0.13.25 internal 1 @noresponse/mcp"
   };
 });
 
@@ -74384,7 +74384,7 @@ endpoint.verification sends a signed challenge now and answers whether the endpo
 function createMCPServer(deps) {
   const server = new McpServer({
     name: "NoResponse",
-    version: "0.13.24"
+    version: "0.13.25"
   });
   const getClient = deps.getSDK || (() => new NoResponseCore({
     security: deps.security,
@@ -83817,7 +83817,7 @@ var routes = buildRouteMap({
 var app = buildApplication(routes, {
   name: "mcp",
   versionInfo: {
-    currentVersion: "0.13.24"
+    currentVersion: "0.13.25"
   }
 });
 run(app, process3.argv.slice(2), buildContext(process3));
@@ -83825,5 +83825,5 @@ export {
   app
 };
 
-//# debugId=1C03D930D6E5233264756E2164756E21
+//# debugId=7E44DD64E08F238864756E2164756E21
 //# sourceMappingURL=mcp-server.js.map