--- id: claude-compliance title: Claude Compliance sidebar_label: Claude Compliance description: The Sumo Logic app for Claude Compliance analyzes activity, security, and compliance data from the Claude platform to support governance and threat detection. slug: /help/docs/integrations/saas-cloud/claude-compliance/ canonical: https://www.sumologic.com/help/docs/integrations/saas-cloud/claude-compliance/ --- import useBaseUrl from '@docusaurus/useBaseUrl'; Claude Compliance icon The Sumo Logic app for Claude Compliance provides security, compliance, and operations teams with centralized visibility into Claude platform and local session activity, covering API usage, authentication events, billing operations, data access, integrations, SSO, policy changes, and Claude local session interactions. It includes prebuilt dashboards for local sessions, messages, and tool activity so teams can monitor user behavior, model usage, workspace context, and tool execution outcomes, alongside detection monitors that help you identify suspicious behavior, investigate threats, and maintain governance across your Claude environment. ## Log types This app uses Claude Compliance's activity logs, chat message logs, and local session endpoint logs. ### Sample log message ```json title="Activity Log" { "actor": { "type": "user_actor", "user_agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 17_1 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) CriOS/120.0.6099.119 Mobile/15E148 Safari/604.1", "user_id": "user_01KQS2GBC9MK53Y7WE6G16G2Y8", "email_address": "alex.martinez482@example.org", "ip_address": "45.150.108.61" }, "id": "activity_01KQS28XP295M0QR9VRCX7C3K0", "type": "org_join_request_instant_approved", "organization_uuid": "org_081m2TJpAuK364YHLwQCXe5r", "created_at": "2026-05-08T12:18:11.000538Z", "organization_id": "01KQS4ZBD35MR4678P2NNXZPW1" } ``` ```json title="Chat Messages Log" { "id": "claude_chat_01BKzmfeRdSfq7otZxLhEAUp", "name": "Greeting", "created_at": "2026-05-07T17:29:59.354315Z", "updated_at": "2026-05-07T17:30:01.282925Z", "deleted_at": null, "organization_id": "org_013w9ZYpBvS942YQNwJAPw6z", "organization_uuid": "17c1b787-c3ac-46e3-a107-0b58fc85b293", "project_id": null, "model": "claude-sonnet-4-6", "user": { "id": "user_01XyDMpzjS89pFZXqSFUBDr6", "email_address": "giacomo@giacomo.plutoenterprise.org" }, "chat_messages": [ { "id": "claude_chat_msg_011Caoc3TT3u8K38ho6SPUPJ", "role": "user", "created_at": "2026-05-07T17:29:59.774439Z", "content": [ { "type": "text", "text": "hey" } ], "files": null, "generated_files": null, "artifacts": null }, { "id": "claude_chat_msg_011Caoc3TT4Gh3TnqUYitFAB", "role": "assistant", "created_at": "2026-05-07T17:30:01.282925Z", "content": [ { "type": "text", "text": " Hey! How can I help you today?" } ], "files": null, "generated_files": null, "artifacts": null } ], "href": "https://claude.ai/chat/53a26065-dc3e-4f8d-99da-5fb6525f14b9", "has_more": false, "first_id": "eyJtc2dfdXVpZCI6ICIwMTllMDM3ZC0zYTRjLTc2NGMtOGMxMy1iZGI0ZDNkZTU4NzkifQ==", "last_id": "eyJtc2dfdXVpZCI6ICIwMTllMDM3ZC0zYTRjLTdjZWEtYjJlNi01M2E5YWNjMDRmYjQifQ==" } ``` ```json title="Local Session Endpoint Log" { "session": { "type": "compliance_local_session", "id": "clls_eyJ2IjoxLCJvIjoiZmJjMWY3MzAtZTBhZi00MGVlLWEwYmUtNjA1N2QxOTcxYjk3IiwicCI6IjhlMWQ5OTNhLTM2NjUtNDc2Ni1hMDUxLTk2YWZjYmIzNDhiMiIsInMiOiI4OWRhZGJlYS1jNDQ5LTQ1YjAtYmJhYy1iZTUzOTI5OWVmM2QifQ", "organization_uuid": "fbc1f730-e0af-40ee-a0be-6057d1971b97", "workspace_id": null, "user": { "id": "user_01JYqynkuhhK4gkr2VXdzqF", "email_address": "giacomo@giacomo.plutoenterprise.org" }, "product_surface": "cowork", "created_at": "2026-09-17T05:50:03.258238Z", "updated_at": "2026-09-17T06:12:27.967457Z", "truncated": false }, "message": { "type": "compliance_local_session_message", "id": "clsm_eyJ2IjoxLCJsIjoibXNnXzAxMUNmOFd4aEpVRHJzSmZiVTlqanJFMyIsInIiOiJhc3Npc3RhbnQiLCJwIjoxLCJxaCI6ImY5ZjdkNmQ1In0", "role": "user", "created_at": "2026-09-17T06:12:27.967457Z", "provenance": null, "model": "claude-sonnet-5", "content": [ { "type": "text", "text": "hey, how are you? Do you know about the claude compliance source in SumoLogic?", "truncated": false } ] } } ``` ### Sample queries ```sumo title="Compliance Activities" _sourceCategory="claude_compliance" type actor | json "type","organization_uuid","id","actor.type" as event_type,org_uuid,id,actor_type nodrop // Global filter | where if("{{event_type}}" = "*",true,event_type matches "{{event_type}}") | where if("{{organization}}" = "*",true,org_uuid matches "{{organization}}") | where if("{{actor_type}}" = "*",true,actor_type matches "{{actor_type}}") // Panel specific | count by id | count ``` ```sumo title="Activity Types" _sourceCategory="claude_compliance" type actor | json "type","organization_uuid","id","actor.type" as event_type,org_uuid,id,actor_type nodrop // Global filter | where if("{{event_type}}" = "*",true,event_type matches "{{event_type}}") | where if("{{organization}}" = "*",true,org_uuid matches "{{organization}}") | where if("{{actor_type}}" = "*",true,actor_type matches "{{actor_type}}") // Panel specific | count by id,event_type | count by event_type | count ``` ```sumo title="Messages by Role" _sourceCategory="claude_compliance" "claude_chat_msg" | json "id", "updated_at", "user.id", "user.email_address", "name", "organization_id", "message.role", "message.content[0].type", "message.content[0].text", "message.model", "message.id" as id, updated_at, user_id, user_email, name, organization_id, role, message_type, message_value, model, message_id nodrop | where user_email matches "{{user}}" or isBlank(user_email) | where role matches "{{role}}" or isBlank(role) | where model matches "{{model}}" or isBlank(model) | where message_type matches "{{message_type}}" or isBlank(message_type) | where message_value contains "{{message}}" or "{{message}}" == "*" | where !isBlank(role) | count by message_id, role | count by role | sort by _count, role ``` ```sumo title="Local Session Messages by Role" _sourceCategory="claude_compliance" "compliance_local_session_message" | json "session.id", "session.product_surface", "session.user.email_address", "message.id", "message.role", "message.model", "message.content[0].type" as session_id, product_surface, user_email, message_id, role, model, message_type nodrop | where user_email matches "{{user}}" or isBlank(user_email) | where role matches "{{role}}" or isBlank(role) | where model matches "{{model}}" or isBlank(model) | where product_surface matches "{{product_surface}}" or isBlank(product_surface) | where !isBlank(role) | count by message_id, role | count by role | sort by _count, role ``` ## Collection configuration This app requires three types of logs: **activity logs** (compliance events across your Claude environment), **chat message logs** (conversation-level data from the Claude Messages API), and **local session endpoint logs** (session, message, and tool activity from Claude local sessions). We recommend using the dedicated Cloud-to-Cloud (C2C) source, which supports collecting all three log types and handles API-specific behaviors automatically. ### Vendor configuration To collect logs, you need a Claude API key with access to the Compliance API. Use one of the following options to create the API key: :::note Admin keys created through Console are limited to the Activity Feed and cannot access chat messages or local session data via C2C source. ::: #### Console / API Keys are created in the **Admin keys** section of Console Settings. 1. Click **Create key** to name your key. 2. Receive a secret access key and store it securely. :::note If the Compliance API is enabled for your organization, Admin keys created here are automatically granted the `read:compliance_activities` scope. If the Compliance API is not yet enabled, contact your Anthropic representative to request access. ::: #### Claude.ai Keys are created in the **Compliance access keys** section of Data Management Settings. 1. Click **Create key** to name your key. 2. Name the key and select its scopes. For activities select `read:compliance_activities`, and for chat messages and local session data select `read:compliance_user_data`. 3. Receive a secret access key and store it securely. :::note If you do not see the Compliance access keys section, either you are not a Primary Owner of the organization, or the Compliance API is not enabled for your organization. The Primary Owner needs to enable it in the Data and Privacy section of your organization's settings. ::: ### Collection methods 1. **[Cloud-to-Cloud (C2C) source](/docs/send-data/hosted-collectors/cloud-to-cloud-integration-framework/claude-compliance-source/) (recommended)**. Supports collection of activity logs, chat message logs, and local session endpoint logs. Handles Anthropic's Compliance API ingestion delay automatically, ensuring no data is missed at polling boundaries. 2. **Universal Connector**. Collects activity logs only — chat message logs and local session endpoint logs are not supported. Because the Universal Connector does not account for Anthropic's ~1-minute API propagation delay, events that arrive at the API boundary may be missed. For this reason, the dedicated C2C source is preferred. ### Details for C2C source (recommended) Use the [Cloud-to-Cloud Integration for Claude Compliance](/docs/send-data/hosted-collectors/cloud-to-cloud-integration-framework/claude-compliance-source/) to create the source. Ensure the `_sourceCategory` configured on your source matches the source category used when installing the app. ### Details for Universal Connector :::warning The Universal Connector collects activity logs only. It does not support chat message logs or local session endpoint logs, and does not account for Anthropic's ~1-minute Compliance API propagation delay, which may cause data to be missed at polling window boundaries. Use the [C2C source](#details-for-c2c-source-recommended) instead. ::: 1. On the Data Collection page, click **Add Source** next to a Hosted Collector. 1. Search for and select **Universal Connector**. 1. Configure the **General** settings: - **Name**. Enter a name for the source. - **Description**. (Optional) Enter a description. - **Source Category**. Enter a value such as `claude_compliance`. This value is stored in the `_sourceCategory` metadata field and must match the source category used when installing the app. - **Fields**. (Optional) Click **+Add** to define any additional fields to associate with the source. Universal Connector - General settings 1. Configure the **Authentication Configuration**: - **Authentication Type**. Select **API Key**. - **How should we use your API key?** Select **In HTTP Request Header**. - **Location Key**. Enter `x-api-key`. - **API Key**. Enter the Claude API key you copied above. Universal Connector - Authentication Configuration 1. Configure the **Request Configuration**: - **HTTP Method**. Select `GET`. - **Endpoint URL**. Enter `https://api.anthropic.com/v1/compliance/activities`. Universal Connector - Request Configuration 1. Configure the **Tracking Progression**: - **Type**. Select **Time Window**. - **Window Size**. Enter `5m` (default recommended). - **Initial Lookback**. Enter `24h`. - **Progress Window Parameters**. Add the following parameters: | Parameter Name | Parameter Value | |:--|:--| | `created_at.gte` | `{{ .WindowStartUTC "yyyy-MM-ddTHH:mm:ssZ" }}` | | `created_at.lt` | `{{ .WindowEndUTC "yyyy-MM-ddTHH:mm:ssZ" }}` | Universal Connector - Tracking Progression 1. Configure the **HTTP Response Log Ingest Configuration**: - **Format**. Select **JSON with JPath**. - Configure the following log path settings: | Field | Value | |:--|:--| | **Logs JPath** | `$.data[*]` | | **Timestamp JPath** | `$.created_at` | | **Timestamp Format** | `2006-01-02T15:04:05.000000Z` | Universal Connector - HTTP Response Log Ingest Configuration 1. Configure the **Pagination Configuration**: - **Type**. Select **Continuation Token**. - **Token Location**. Select **Body**. - **Next Page Continuation Token JPath**. Enter `$.last_id`. - **Send Token In**. Select **Parameters**. - **Parameter Key**. Enter `after_id`.
Universal Connector - Pagination Configuration 1. (Optional) Configure the **HTTP Client Configuration**: - **HTTP Timeout**. `5m` (default). - **HTTP Client Retries**. `5` (default). - **Rate Limit Requests**. `1000` (default). - **Rate Limit Duration**. `1m` (default). - **Rate Limit Burst**. `1000` (default). - **Polling Interval**. Set how frequently to poll for new data, between 5 minutes and 48 hours. Universal Connector - HTTP Client Configuration 1. Click **Save**. :::note Once the source is configured, you can verify successful log collection by running searches on the Search page in Sumo Logic using the source category. For example, `_sourceCategory=claude_compliance`. ::: ## App installation import CollectionConfiguration from '../../reuse/apps/collection-configuration.md'; ### Create a new collector and install the app import AppCollectionOPtion1 from '../../reuse/apps/app-collection-option-1.md'; ### Use an existing collector and install the app import AppCollectionOPtion2 from '../../reuse/apps/app-collection-option-2.md'; ### Use an existing source and install the app import AppCollectionOPtion3 from '../../reuse/apps/app-collection-option-3.md'; ## Viewing the Claude Compliance dashboards​​ import ViewDashboards from '../../reuse/apps/view-dashboards.md'; ### Activity Overview The **Claude Compliance - Activity Overview** dashboard delivers a high-level summary of all compliance-relevant activities occurring across your Claude environment, including event volumes, activity types, and threat counts. It maps activity by geographic location, highlights events from embargoed regions, and identifies unauthenticated actors and failed activities. Teams can drill into activity distribution by organization, actor, and identity provider connection type. **Use this dashboard to:** - Quickly assess the overall compliance posture. - Surface anomalous behavior patterns. - Detect activities from embargoed regions. Claude Compliance - Activity Overview dashboard ### Billing Monitoring The **Claude Compliance - Billing Monitoring** dashboard monitors billing-related compliance events, including subscription lifecycle changes, spend limit updates, prepaid activity, and recharge events across your organization. It highlights the top plan types, organizations with the most billing activity, and users making frequent spend limit changes. Trend analysis and detailed event tables provide full audit coverage for financial governance and compliance. **Use this dashboard to:** - Detect unauthorized billing changes. - Track subscription modifications. - Ensure spend controls are functioning as intended. Claude Compliance - Billing Monitoring dashboard ### API Key Monitoring The **Claude Compliance - API Key Monitoring** dashboard provides visibility into API key lifecycle events across your Claude environment, including creation, deletion, and modification of API, Admin, Service, and Signing keys. It tracks key event trends over time, highlights the most active users, and surfaces detailed event records for audit and investigation purposes. **Use this dashboard to:** - Detect unauthorized key activity. - Monitor privileged admin key usage. - Ensure key management practices align with your security policies. Claude Compliance - API Key Monitoring dashboard ### Artifacts, Marketplace, and Projects Monitoring The **Claude Compliance - Artifacts, Marketplace, and Projects Monitoring** dashboard tracks lifecycle events for Artifacts, Marketplace items, and Projects on the Claude platform, including creation, access, and deletion. It provides trend analysis, event breakdowns, and detailed activity tables to help teams identify unusual access patterns or unauthorized changes. Active artifact, project, and user counts give a real-time snapshot of platform engagement. **Use this dashboard to:** - Maintain governance over shared resources. - Ensure marketplace and project activities align with organizational policies. Claude Compliance - Artifacts, Marketplace, and Projects Monitoring dashboard ### Claude Skills and Plugins Monitoring The **Claude Compliance - Claude Skills and Plugins Monitoring** dashboard provides visibility into the usage and lifecycle of Claude Skills and Plugins across your Claude environment, including invocation trends, event distributions, and top active skills and plugins. It surfaces unique skill and plugin counts, along with detailed activity records, to help teams govern AI capability extensions. Usage trends help identify sudden spikes or drops in plugin or skill activity that may warrant investigation. **Use this dashboard to:** - Ensure approved extensions are being used appropriately. - Detect unauthorized or anomalous AI tool usage. Claude Compliance - Claude Skills and Plugins Monitoring dashboard ### Claude Usage Monitoring The **Claude Compliance - Claude Usage Monitoring** dashboard delivers comprehensive visibility into Claude platform usage, covering chat events, chat snapshots, Claude Code activity, and Claude Command executions. It tracks usage trends, top active users, unique interaction counts, failed chats, and model usage distribution to support both operational monitoring and compliance oversight. Dedicated sections for Claude Code provide insight into trigger modes, top repositories, and event statuses across development workflows. **Use this dashboard to:** - Enforce acceptable use policies. - Detect excessive or anomalous usage. - Maintain a complete audit trail of Claude's interactions. Claude Compliance - Claude Usage Monitoring dashboard ### Compliance API Access Monitoring The **Claude Compliance - Compliance API Access Monitoring** dashboard monitors access to Claude's Compliance API, providing visibility into request trends by user agent, HTTP status code distributions, and geographic origin of API calls. It highlights non-2xx error trends, top actor IP addresses, and the types of API endpoints accessed to help teams detect misuse or unauthorized access attempts. It provides geo-location filtering and surfaced API access from restricted regions for immediate review. **Use this dashboard to:** - Ensure Compliance API access is limited to authorized actors. - Investigate anomalous or error-prone API usage patterns. Claude Compliance - Compliance API Access Monitoring dashboard ### Data Access and Exfiltration Monitoring The **Claude Compliance - Data Access and Exfiltration Monitoring** dashboard provides deep visibility into file, document, and data export activities across the Claude platform helping detect potential data exfiltration or unauthorized access. It tracks file access events, reconnaissance behavior, Claude artifact sharing, and platform-level file activities alongside group and project views. Frequently uploaded and most-viewed file analysis helps identify unusual data movement patterns. **Use this dashboard to:** - Monitor data governance compliance. - Investigate suspicious exfiltration indicators. - Maintain oversight of sensitive content access and sharing. Claude Compliance - Data Access and Exfiltration Monitoring dashboard ### Identity, Access, and Authentication Monitoring The **Claude Compliance - Identity, Access, and Authentication Monitoring** dashboard provides comprehensive monitoring of authentication events, role and group changes, SCIM provisioning, and API key activity across your Claude environment. It tracks login success and failure trends, highlights users with failed authentication, and maps activity by geographic location, including embargoed regions. Privilege escalation flows and recon-exploit patterns are surfaced to help security teams identify and respond to identity-based threats. **Use this dashboard to:** - Enforce least-privilege access. - Investigate authentication anomalies. - Maintain a complete audit trail of identity and access events. Claude Compliance - Identity, Access, and Authentication Monitoring dashboard ### MCP Servers, Integrations, and GitHub Enterprise The **Claude Compliance - MCP Servers, Integrations, and GitHub Enterprise** dashboard monitors management events for MCP Servers, third-party Integrations, and GitHub Enterprise connections within your Claude environment. It tracks active MCP servers, integration types, top GitHub repositories, and the geo-locations of actors performing management actions. Event distribution and detail tables provide full visibility into configuration changes and integration lifecycle activity. **Use this dashboard to:** - Govern AI infrastructure extensions. - Detect unauthorized changes to integrations. - Ensure GitHub Enterprise connectivity aligns with organizational security standards. Claude Compliance - MCP Servers, Integrations, and GitHub Enterprise dashboard ### Organization and Policy Configuration The **Claude Compliance - Organization and Policy Configuration** dashboard monitors organization-level configuration events, including discoverability changes, org management actions, disabled events, and user policy modifications. It maps the geo-location of users performing org changes, highlights top users making configuration updates, and tracks event trends over time. Detailed event tables and breakdowns support rapid investigation of unauthorized or unexpected organizational changes. **Use this dashboard to:** - Ensure configuration governance. - Detect policy drift. - Maintain an auditable record of all organization-level administrative activity. Claude Compliance - Organization and Policy Configuration dashboard ### SSO, Directory Sync Monitoring The **Claude Compliance - SSO, Directory Sync Monitoring** dashboard tracks Single Sign-On (SSO) and Directory Sync events across your Claude environment, including connection activations, deactivations, deletions, and sync lifecycle activity. It provides event distributions, trend analysis, and geographic mapping of SSO and directory sync activities to support identity governance. Top users performing SSO and directory sync changes are highlighted for accountability and anomaly detection. **Use this dashboard to:** - Ensure SSO configurations remain secure. - Detect unauthorized changes to identity providers. - Maintain compliance with access management policies. Claude Compliance - SSO, Directory Sync Monitoring dashboard ### Session Management The **Claude Compliance - Session Management** dashboard monitors user session activity across the Claude platform, including session creation, sharing, and termination events. It visualizes session event distributions, activity trends, and geographic locations of session activity, including events from embargoed regions. A top user's view highlights accounts with the highest session activity for anomaly detection and access governance. **Use this dashboard to:** - Detect suspicious session behavior. - Enforce session policy compliance. - Investigate unauthorized or geographically anomalous session access. Claude Compliance - Session Management dashboard ### User Agent Analysis The **Claude Compliance - User Agent Analysis** dashboard analyzes the user agents accessing your Claude environment, categorizing traffic by browser, operating system, platform, and automated versus human origin. It tracks user-agent activity trends, surfaces the top agents associated with failed activities, and highlights HTTP errors encountered by different client types. Connector and category trend analysis helps teams identify unusual or unauthorized client tooling over time. **Use this dashboard to:** - Detect bot activity. - Investigate suspicious client behavior. - Ensure that only approved tools and platforms access your Claude environment. Claude Compliance - User Agent Analysis dashboard ### User Configuration and Invite Lifecycle Monitoring The **Claude Compliance - User Configuration and Invite Lifecycle Monitoring** dashboard provides full visibility into user configuration changes, invite lifecycle events, role assignments, and organizational membership activity across your Claude environment. It tracks users with the highest frequency of configuration changes, monitors invite link usage, and surfaces geolocation data on user activities, including access from embargoed regions. Detailed tables cover role changes, org member invites, user settings updates, and recent activity for comprehensive audit coverage. **Use this dashboard to:** - Govern user onboarding and offboarding processes. - Detect unauthorized role changes. - Ensure invite and configuration practices comply with organizational policies. Claude Compliance - User Configuration and Invite Lifecycle Monitoring dashboard ### Chats The **Claude Compliance – Chats** dashboard provides comprehensive oversight of Claude AI usage across your organization, surfacing key metrics including active users, deployed models, project distribution, and message volume trends. It breaks down interactions by role and content type, ranks top models and users, and provides audit-ready tables for generated artifacts, file uploads, and conversation summaries. Granular message-level data supports deep compliance review and sensitive data detection. **Use this dashboard to:** - Ensure transparency and responsible AI use across your organization. - Support deep compliance review and sensitive data detection. - Maintain data governance compliance. :::note Privacy Protection Panels with sensitive conversation data require applying the "message" filter with a specified text value to view detailed results, ensuring controlled access to potentially sensitive information. ::: :::note This dashboard is powered by Sumo Logic [Claude Compliance Source](/docs/send-data/hosted-collectors/cloud-to-cloud-integration-framework/claude-compliance-source/). ::: Claude Compliance - Chats dashboard ### Tools The **Claude Compliance - Tools** dashboard provides visibility into tool usage within Claude local sessions, including invocation volume, failed tool executions, error rate, block-type distribution, and frequently used tools. It surfaces recent tool activity, tool response details, and user-level usage patterns to help teams audit how tools are being used during Claude interactions. **Use this dashboard to:** - Identify failing, risky, or unexpected tool behavior across local workflows. - Audit tool invocation volume and error rates. - Review recent tool activity and response details by user. :::note This dashboard is powered by Sumo Logic [Claude Compliance Source](/docs/send-data/hosted-collectors/cloud-to-cloud-integration-framework/claude-compliance-source/). ::: Claude Compliance - Tools dashboard ### Messages The **Claude Compliance - Messages** dashboard provides message-level visibility into Claude local session activity across users, models, organizations, and product surfaces. It tracks message volume, role distribution, model usage trends, provenance states, truncated content, and content_unavailable events to support audit and investigation workflows. Search and drill-down panels help teams review message activity, investigate specific keywords, and identify unusual or risky usage patterns across local Claude interactions. **Use this dashboard to:** - Investigate message content and keywords across local sessions. - Track model usage trends and role distribution. - Identify unusual or risky usage patterns. :::note This dashboard is powered by Sumo Logic [Claude Compliance Source](/docs/send-data/hosted-collectors/cloud-to-cloud-integration-framework/claude-compliance-source/). ::: Claude Compliance - Messages dashboard ### Sessions The **Claude Compliance - Sessions** dashboard provides a session-level view of Claude local session activity across users, workspaces, organizations, and product surfaces. It tracks active sessions, unique users, workspace distribution, average session span, and session creation trends to help teams understand where and how Claude is being used locally. Recent session details and activity breakdowns support governance, usage auditing, and investigation of anomalous session behavior. **Use this dashboard to:** - Understand where and how Claude is being used locally. - Audit session creation trends and workspace distribution. - Investigate anomalous session behavior. :::note This dashboard is powered by Sumo Logic [Claude Compliance Source](/docs/send-data/hosted-collectors/cloud-to-cloud-integration-framework/claude-compliance-source/). ::: Claude Compliance - Sessions dashboard ## Create monitors for Claude Compliance app import CreateMonitors from '../../reuse/apps/create-monitors.md'; ### Claude Compliance monitors | Name | Description | Trigger Type | Alert Condition | |:--|:--|:--|:--| | `Claude Compliance - Admin or Platform API Key Created` | Detects when an Admin API key or Platform API key is created within the organization. Adversaries or malicious insiders may create API keys to establish persistent, programmatic access to organizational resources. Any key creation event should be reviewed to verify that it was authorized and expected. | Critical | Count > 0 | | `Claude Compliance - API Logging Disabled` | Detects when Compliance API logging is disabled for an organization. Disabling audit logging is a common defense evasion technique — attackers may disable it to reduce visibility into their activities. Any disabling of compliance API logging should be treated as a high priority and investigated immediately. | Critical | Count > 0 | | `Claude Compliance - Claude Organization IP Restriction Deleted` | Detects when an IP restriction policy is deleted from an organization. IP allowlisting is a critical access control. Removing IP restrictions may allow an attacker to access Claude resources from previously blocked networks, significantly expanding the attack surface. | Critical | Count > 0 | | `Claude Compliance - Compliance Activity From Embargoed Location` | Detects compliance-related activities originating from embargoed or geographically restricted locations. Alerts are triggered when compliance actions are performed from regions that are prohibited under regulatory or organizational guidelines. | Critical | Count > 0 | | `Claude Compliance - Excessive Failed Authentication User Activities Detected` | Detects an abnormal number of failed authentication attempts across multiple authentication mechanisms, including SSO failures and magic link login failures within the Claude platform. Alerts are triggered when failed authentication attempts exceed the threshold within a specific time window, potentially indicating brute force attacks, credential stuffing, or account takeover attempts. | Critical | Count > 5 | | `Claude Compliance - Excessive Unauthorized Access Attempts on Compliance API` | Detects repeated unauthorized and forbidden access attempts on the Claude Compliance API within a short time window. Alerts are triggered when multiple 401 Unauthorized or 403 Forbidden responses are observed from the same user or IP address, indicating potential credential abuse, brute force attempts, or unauthorized API access activity. | Critical | Count > 5 | | `Claude Compliance - Org Deletion & Destructive Activity` | Detects any destructive actions, including organization deletions, bulk deletes, user removals, and taint additions within the Claude platform. Alerts are triggered when irreversible deletion or destructive events are performed, indicating potential unauthorized data destruction or malicious insider activity requiring immediate investigation. | Critical | Count > 0 | | `Claude Compliance - Outlier in Claude Resource Deletions from User` | Detects unusual spikes in resource deletion activity performed by a single user within the Claude platform. Alerts are triggered when a user's deletion activity significantly deviates from their historical baseline, indicating potential insider threat, compromised account, or unauthorized mass deletion of critical resources. | Critical | Count > 1 | | `Claude Compliance - Unauthenticated User Activity Detected` | Detects suspicious activities or access attempts made by unauthenticated users within the Claude platform. Alerts are triggered when restricted resources are accessed without valid authentication or when unusual patterns are identified in unauthenticated traffic. | Critical | Count > 0 | ## Upgrade/Downgrade the Claude Compliance app (Optional) import AppUpdate from '../../reuse/apps/app-update.md'; ## Uninstalling the Claude Compliance app (Optional) import AppUninstall from '../../reuse/apps/app-uninstall.md';