{"generated_at":"2026-10-14T20:07:22+00:00","item":{"added_at":"2026-06-24","analysis_method":"independent_manual_review_2026-10-10","assessment_status":"provisional","auth_posture":"auth_required","category":"Web Connector","confidence":"medium","description":"Investigate security data and events from Exaforce in Claude. Search users, endpoints and alerts and correlate activity with configuration. The connector also supports case updates, rules, automation and endpoint-response actions.","did":"14de6d99-a0ef-490a-85f5-2d10dd3bc424","documentation_url":"https://www.exaforce.com/platform/exabot-investigate","editorial_description":{"extended_description":"Coverage follows the existing Exaforce ingestion pipeline and connected sources. Listed actions include creating cases, changing alert status, isolating a SentinelOne device and starting agents. Access is scoped to the authenticated tenant; some tools also query external security services.","schema_version":1,"summary":"Investigate security data and events from Exaforce in Claude.\nSearch users, endpoints and alerts and correlate activity with configuration.\nThe connector also supports case updates, rules, automation and endpoint-response actions."},"homepage_url":"","icon":"https://cdn.prod.website-files.com/67d895a4a98745273854676e/689044a050385aae409a06ea_Exaforce-Website-Favicon.png","id":"14de6d99-a0ef-490a-85f5-2d10dd3bc424","installs":0,"last_scanned":"","license":"","long_description":"Coverage follows the existing Exaforce ingestion pipeline and connected sources. Listed actions include creating cases, changing alert status, isolating a SentinelOne device and starting agents. Access is scoped to the authenticated tenant; some tools also query external security services.","manual_assessment":"high","manual_review":{"all_findings":[{"evidence":"The inventory explicitly includes execution of Defender hunting, BigQuery audit and generic queries.","reasoning":"These tools run expressive query languages over sensitive security and audit datasets.","risk_type":"query_execution","severity":"high","title":"Executes security hunting and tenant queries","triggering_tools_or_skills":["DefenderXDRAdvancedHuntingQuery","ExecuteQuery","GoogleWorkspaceQueryAuditReportsBigquery","QueryBuilderRequestExecuteQuery"]},{"evidence":"The declared tools isolate an endpoint and update security-control records.","reasoning":"Endpoint isolation is a powerful security-control action that can disrupt production access.","risk_type":"credential_admin","severity":"high","title":"Isolates devices and changes security alert and threat state","triggering_tools_or_skills":["SentinelOneIsolateDevice","SentinelOneUpdateAlertStatus","SentinelOneUpdateThreatStatus"]},{"evidence":"The inventory explicitly deletes a persistent business-context rule.","reasoning":"This removes security-analysis configuration.","risk_type":"data_deletion","severity":"high","title":"Deletes business-context rules","triggering_tools_or_skills":["ManageBusinessContextRuleDeleteRule"]},{"evidence":"The connector can discover agent inputs, start an agent and retrieve its result, without enumerating the agent action set.","reasoning":"Agent execution can encapsulate additional security operations not visible in the static inventory.","risk_type":"generic_gateway","severity":"high","title":"Starts automation agents with runtime parameters","triggering_tools_or_skills":["GetAgentInputParameters","StartAgent","GetAutomationAgentResult"]},{"evidence":"The connector retrieves collaboration pages, issues, external search results and scanned URL content.","reasoning":"External and user-authored text can contain adversarial instructions.","risk_type":"prompt_injection_exposure","severity":"low","title":"Tickets, pages and security telemetry contain untrusted text","triggering_tools_or_skills":["ConfluenceGetPage","JiraGetIssue","JiraSearchIssues","PerplexitySearch","UrlScanTool"]}],"assessment_status":"provisional","auth_and_scopes":"OAuth 2.1 with PKCE scopes access to an Exaforce tenant, but query languages, endpoint-isolation privileges, security-case roles and agent actions are not documented.","blocked_missing_evidence":null,"confidence":"medium","displayed_risk":"high","evidence_completeness":"partial","manual_overall_risk":"high","official_sources":[{"accessed_at":"2026-10-10T20:24:37Z","url_or_path":"plutosecurity/plutonium@b749c8dc43536c794409be9805edd48063f045e2:app/data/conn/claudesec/14de6d99-a0ef-490a-85f5-2d10dd3bc424.json","what_it_proves":"Candidate PR #125 record identity (name \"Exaforce\", publisher, type web_connector) and 42 stored tool names; no tool descriptions stored"},{"accessed_at":"2026-09-22T10:46:07Z","url_or_path":"https://api.anthropic.com/api/directory/servers?verified_tier=anthropic,partner,community (snapshot claude-directory-all-tiers-2026-09-22.json)","what_it_proves":"Official Claude Directory listing https://claude.ai/directory/14de6d99-a0ef-490a-85f5-2d10dd3bc424: complete tool_names inventory (42 tools), remote streamable-http endpoint https://mcp.api.us.app.exaforce.io/, auth_posture auth_required, directory tier community"}],"primary_reason":"The connector executes security queries, isolates endpoints, deletes rules and can launch opaque automation agents.","provisional_risk":null,"publication_recommendation":"fix","requires_runtime_verification":true,"review_key":"claude:14de6d99-a0ef-490a-85f5-2d10dd3bc424","reviewed_at":"2026-10-10","reviewer_notes":"Reviewed all 42 names. Publish the StartAgent catalogue, query schemas and role requirements, and verify URL-scan target restrictions.","risk_domains":{"arbitrary_network_access":"true","authz_ambiguity":"true","browser_control":"false","credential_or_admin_actions":"true","deletes_data":"true","deployment_or_infrastructure":"false","dynamic_hidden_actions":"true","filesystem_or_repo_access":"false","forwards_data_to_third_party":"true","modifies_data":"true","persistent_credentials":"true","prompt_injection_exposure":"true","provenance_uncertainty":"true","reads_private_data":"true","runs_code_or_queries":"true","sends_messages_as_user":"false","spends_or_moves_money":"false"},"tools_or_skills_reviewed":42,"tools_or_skills_total":42},"name":"Exaforce","needs_runtime_verification":true,"plutonium_url":"https://plutonium.pluto.security/detail.html?planet=claudesec&did=14de6d99-a0ef-490a-85f5-2d10dd3bc424&utm_source=plutonium_analysis_skill&utm_medium=claude_skill&utm_campaign=connector_risk_assessment","privacy_policy_url":"https://www.exaforce.com/policies/privacy-policy","provisional_risk":null,"publication_recommendation":"fix","publisher":"Exaforce","publisher_url":"https://exaforce.com/","remote_url":"https://mcp.api.us.app.exaforce.io/","repository_url":"","risk":"high","risk_severity":"high","security_risks":[{"description":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory.","evidence":"Executes security hunting and tenant queries: The inventory explicitly includes execution of Defender hunting, BigQuery audit and generic queries. | Starts automation agents with runtime parameters: The connector can discover agent inputs, start an agent and retrieve its result, without enumerating the agent action set.","manual_risk_type":"query_execution,generic_gateway","remediation":{"block_tool_categories":["code_execution"],"steps":["Run only in a sandbox with least-privilege credentials and strict resource limits.","Disable generic execution gateways unless their reachable action set is fully understood.","Require approval before execution against production systems or sensitive data."]},"risk_type":"runs_code","severity":"high","title":"Can execute code, queries, or dynamic actions","triggering_tools":["DefenderXDRAdvancedHuntingQuery","ExecuteQuery","GoogleWorkspaceQueryAuditReportsBigquery","QueryBuilderRequestExecuteQuery","GetAgentInputParameters","StartAgent","GetAutomationAgentResult"]},{"description":"This connector can create, update, configure, or otherwise change connected resources. Review findings: Endpoint isolation is a powerful security-control action that can disrupt production access.","evidence":"Isolates devices and changes security alert and threat state: The declared tools isolate an endpoint and update security-control records.","manual_risk_type":"credential_admin","remediation":{"block_tool_categories":["state_change"],"steps":["Use a test account or sandbox before allowing changes to production data.","Grant only the write permissions required for the intended workflow.","Require human confirmation for consequential state changes."]},"risk_type":"modifies_data","severity":"high","title":"Isolates devices and changes security alert and threat state","triggering_tools":["SentinelOneIsolateDevice","SentinelOneUpdateAlertStatus","SentinelOneUpdateThreatStatus"]},{"description":"This connector can permanently remove data or resources. Review findings: This removes security-analysis configuration.","evidence":"Deletes business-context rules: The inventory explicitly deletes a persistent business-context rule.","manual_risk_type":"data_deletion","remediation":{"block_tool_categories":["destructive"],"steps":["Disable destructive tools unless deletion is an explicit workflow requirement.","Require human confirmation and verified backups before deletion."]},"risk_type":"deletes_data","severity":"high","title":"Deletes business-context rules","triggering_tools":["ManageBusinessContextRuleDeleteRule"]}],"signature_status":"unknown","source_code_reviewed":false,"support_url":"support@exaforce.com","tags":["deletes_data","modifies_data","network_access","oauth_required","runs_code","web_connector"],"tools":[{"description":"","name":"CQLGeneration"},{"description":"","name":"ConfluenceGetPage"},{"description":"","name":"DatabaseLookup"},{"description":"","name":"DefenderXDRAdvancedHuntingQuery","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory."}},{"description":"","name":"DefenderXDRGetIncident"},{"description":"","name":"DesignTemplateGetTemplate"},{"description":"","name":"DesignTemplateListTemplates"},{"description":"","name":"ExecuteQuery","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory."}},{"description":"","name":"GeoIPLookupTool"},{"description":"","name":"GetAgentInputParameters","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory."}},{"description":"","name":"GetAutomationAgentResult","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory."}},{"description":"","name":"GoogleWorkspaceListOauthTokens"},{"description":"","name":"GoogleWorkspaceQueryAuditReportsBigquery","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory."}},{"description":"","name":"JiraGetIssue"},{"description":"","name":"JiraSearchIssues"},{"description":"","name":"ListAvailableQuestions"},{"description":"","name":"ListQuestionsDetails"},{"description":"","name":"LookUpTableGetData"},{"description":"","name":"ManageBusinessContextRuleCreateRule"},{"description":"","name":"ManageBusinessContextRuleDeleteRule","risk":{"category":"destructive","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can permanently remove data or resources. Review findings: This removes security-analysis configuration."}},{"description":"","name":"ManageBusinessContextRuleFindSimilarRules"},{"description":"","name":"ManageBusinessContextRuleGetRule"},{"description":"","name":"ManageBusinessContextRuleListRules"},{"description":"","name":"ManageBusinessContextRuleUpdateRule"},{"description":"","name":"ManageCaseAddNote"},{"description":"","name":"ManageCaseCreateCase"},{"description":"","name":"ManageCaseGetCase"},{"description":"","name":"ManageCaseUpdateCase"},{"description":"","name":"PerplexitySearch"},{"description":"","name":"QueryBuilderRequestBuildQuery"},{"description":"","name":"QueryBuilderRequestBuildQueryV3"},{"description":"","name":"QueryBuilderRequestExecuteQuery","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory."}},{"description":"","name":"SchemaDescribe"},{"description":"","name":"SentinelOneInvestigate"},{"description":"","name":"SentinelOneIsolateDevice","risk":{"category":"state_change","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can create, update, configure, or otherwise change connected resources. Review findings: Endpoint isolation is a powerful security-control action that can disrupt production access."}},{"description":"","name":"SentinelOneUpdateAlertStatus","risk":{"category":"state_change","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can create, update, configure, or otherwise change connected resources. Review findings: Endpoint isolation is a powerful security-control action that can disrupt production access."}},{"description":"","name":"SentinelOneUpdateThreatStatus","risk":{"category":"state_change","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can create, update, configure, or otherwise change connected resources. Review findings: Endpoint isolation is a powerful security-control action that can disrupt production access."}},{"description":"","name":"StartAgent","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: These tools run expressive query languages over sensitive security and audit datasets.; Agent execution can encapsulate additional security operations not visible in the static inventory."}},{"description":"","name":"UrlScanTool"},{"description":"","name":"VirusTotalDomainScan"},{"description":"","name":"VirusTotalFileScan"},{"description":"","name":"VirusTotalUrlScan"}],"tools_count":42,"transport":"streamable-http","type":"web_connector","url":"https://claude.ai/directory/14de6d99-a0ef-490a-85f5-2d10dd3bc424","uuid":"b7482293-6a78-55ae-beb5-9899f3ec62f4","verification_tier":"community","version":""},"query_key":"exaforce","schema_version":1,"status":"match"}