{"generated_at":"2026-10-14T20:07:22+00:00","item":{"added_at":"2026-06-30","analysis_method":"independent_manual_review_2026-10-10","assessment_status":"provisional","auth_posture":"auth_required","category":"Web Connector","confidence":"medium","description":"Sequentum connects Claude to web-data agents and their execution workflows. Build, run, schedule, and stop agents, then inspect results, files, failures, and diagnostics. Review usage, credit balances, and spending across agents and spaces.","did":"4a77f913-6cde-4b5d-9ab1-f83971992c93","documentation_url":"https://docs.sequentum.com/mcp/overview","editorial_description":{"extended_description":"Sequentum's platform supports web scraping, crawling, search, browser orchestration, and web automation. The connector also lists agent versions, restores versions, manages schedules, and retrieves downloadable output files and run cost breakdowns.","schema_version":1,"summary":"Sequentum connects Claude to web-data agents and their execution workflows.\nBuild, run, schedule, and stop agents, then inspect results, files, failures, and diagnostics.\nReview usage, credit balances, and spending across agents and spaces."},"homepage_url":"","icon":"","id":"4a77f913-6cde-4b5d-9ab1-f83971992c93","installs":0,"last_scanned":"","license":"","long_description":"Sequentum's platform supports web scraping, crawling, search, browser orchestration, and web automation. The connector also lists agent versions, restores versions, manages schedules, and retrieves downloadable output files and run cost breakdowns.","manual_assessment":"high","manual_review":{"all_findings":[{"evidence":"The description explicitly provides browser orchestration, web automation, crawling and agentic checkout flows.","reasoning":"Agents interact with arbitrary websites and may perform actions rather than simple reads.","risk_type":"browser_control","severity":"high","title":"Runs browser orchestration and web automation agents","triggering_tools_or_skills":["start_agent","run_space_agents"]},{"evidence":"Natural-language prompts create production-ready agents that are built and run remotely.","reasoning":"The platform executes dynamic extraction/automation programs.","risk_type":"code_execution","severity":"high","title":"Builds and executes production web-data agents","triggering_tools_or_skills":["start_agent_build","start_agent","run_space_agents"]},{"evidence":"The surface exposes credit balance/history, agent cost breakdown and spending summaries alongside run/build starts.","reasoning":"Execution consumes metered paid credits.","risk_type":"financial_transaction","severity":"high","title":"Starts credit-consuming agents and builds","triggering_tools_or_skills":["start_agent","start_agent_build","run_space_agents"]},{"evidence":"The connector manages production scheduled automation runs.","reasoning":"Schedules deploy recurring browser/data workflows that run independently.","risk_type":"deployment_infrastructure","severity":"high","title":"Creates and enables recurring agent schedules","triggering_tools_or_skills":["create_agent_schedule","enable_agent_schedule","update_agent_schedule"]},{"evidence":"The complete inventory includes explicit schedule and run deletion.","reasoning":"These actions remove automation configuration or operational audit/history.","risk_type":"data_deletion","severity":"high","title":"Deletes agent schedules and run records","triggering_tools_or_skills":["delete_agent_schedule","delete_run"]}],"assessment_status":"provisional","auth_and_scopes":"Authentication exposes enterprise web agents, builds, schedules, runs, files, diagnostics, credits and spend summaries.","blocked_missing_evidence":null,"confidence":"medium","displayed_risk":"high","evidence_completeness":"complete","manual_overall_risk":"high","official_sources":[{"accessed_at":"2026-10-10T20:24:37Z","url_or_path":"plutosecurity/plutonium@b749c8dc43536c794409be9805edd48063f045e2:app/data/conn/claudesec/4a77f913-6cde-4b5d-9ab1-f83971992c93.json","what_it_proves":"Candidate PR #125 record identity (name \"Sequentum MCP\", publisher, type web_connector) and 39 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/4a77f913-6cde-4b5d-9ab1-f83971992c93: complete tool_names inventory (39 tools), remote streamable-http endpoint https://mcp.sequentum.com/mcp, auth_posture auth_required, publisher-declared permissions text, declared sensitive data types, directory tier community"}],"primary_reason":"Sequentum builds/runs paid browser agents, deploys schedules and deletes run/schedule state.","provisional_risk":null,"publication_recommendation":"accept","requires_runtime_verification":true,"review_key":"claude:4a77f913-6cde-4b5d-9ab1-f83971992c93","reviewed_at":"2026-10-10","reviewer_notes":"Require URL/action/cost preview, network egress controls and confirmation for checkout-capable automations.","risk_domains":{"arbitrary_network_access":"true","authz_ambiguity":"true","browser_control":"true","credential_or_admin_actions":"false","deletes_data":"true","deployment_or_infrastructure":"true","dynamic_hidden_actions":"true","filesystem_or_repo_access":"true","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":"unknown","spends_or_moves_money":"true"},"tools_or_skills_reviewed":39,"tools_or_skills_total":39},"name":"Sequentum MCP","needs_runtime_verification":true,"plutonium_url":"https://plutonium.pluto.security/detail.html?planet=claudesec&did=4a77f913-6cde-4b5d-9ab1-f83971992c93&utm_source=plutonium_analysis_skill&utm_medium=claude_skill&utm_campaign=connector_risk_assessment","privacy_policy_url":"https://www.sequentum.com/privacy-policy","provisional_risk":null,"publication_recommendation":"accept","publisher":"Sequentum","publisher_url":"https://www.sequentum.com/","remote_url":"https://mcp.sequentum.com/mcp","repository_url":"","risk":"high","risk_severity":"high","security_risks":[{"description":"This connector can navigate pages and interact with browser content on the user's behalf. Review findings: Agents interact with arbitrary websites and may perform actions rather than simple reads.","evidence":"Runs browser orchestration and web automation agents: The description explicitly provides browser orchestration, web automation, crawling and agentic checkout flows.","manual_risk_type":"browser_control","remediation":{"block_tool_categories":["state_change"],"steps":["Use an isolated browser profile without unrelated authenticated sessions.","Require confirmation before form submission, download, upload, or purchase actions."]},"risk_type":"browser_control","severity":"high","title":"Runs browser orchestration and web automation agents","triggering_tools":["start_agent","run_space_agents"]},{"description":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: The platform executes dynamic extraction/automation programs.","evidence":"Builds and executes production web-data agents: Natural-language prompts create production-ready agents that are built and run remotely.","manual_risk_type":"code_execution","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":"Builds and executes production web-data agents","triggering_tools":["start_agent_build","start_agent","run_space_agents"]},{"description":"This connector can spend money, consume paid credits, or change financially consequential resources. Review findings: Execution consumes metered paid credits.","evidence":"Starts credit-consuming agents and builds: The surface exposes credit balance/history, agent cost breakdown and spending summaries alongside run/build starts.","manual_risk_type":"financial_transaction","remediation":{"block_tool_categories":["financial"],"steps":["Set provider-side spending limits and require approval for every financial action.","Use a dedicated low-balance account and review transaction logs."]},"risk_type":"spends_money","severity":"high","title":"Starts credit-consuming agents and builds","triggering_tools":["start_agent","start_agent_build","run_space_agents"]},{"description":"This connector can deploy code or change infrastructure and production environments. Review findings: Schedules deploy recurring browser/data workflows that run independently.","evidence":"Creates and enables recurring agent schedules: The connector manages production scheduled automation runs.","manual_risk_type":"deployment_infrastructure","remediation":{"block_tool_categories":["code_execution","state_change"],"steps":["Default to a sandbox or staging environment.","Require code review, human approval, audit logging, and a tested rollback path for production changes."]},"risk_type":"code_deployment","severity":"high","title":"Creates and enables recurring agent schedules","triggering_tools":["create_agent_schedule","enable_agent_schedule","update_agent_schedule"]},{"description":"This connector can permanently remove data or resources. Review findings: These actions remove automation configuration or operational audit/history.","evidence":"Deletes agent schedules and run records: The complete inventory includes explicit schedule and run deletion.","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 agent schedules and run records","triggering_tools":["delete_agent_schedule","delete_run"]}],"signature_status":"unknown","source_code_reviewed":false,"support_url":"https://www.sequentum.com/resources","tags":["browser_control","code_deployment","deletes_data","network_access","oauth_required","runs_code","spends_money","web_connector"],"tools":[{"description":"","name":"create_agent_schedule","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can deploy code or change infrastructure and production environments. Review findings: Schedules deploy recurring browser/data workflows that run independently."}},{"description":"","name":"delete_agent_schedule","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: These actions remove automation configuration or operational audit/history."}},{"description":"","name":"delete_run","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: These actions remove automation configuration or operational audit/history."}},{"description":"","name":"disable_agent_schedule"},{"description":"","name":"enable_agent_schedule","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can deploy code or change infrastructure and production environments. Review findings: Schedules deploy recurring browser/data workflows that run independently."}},{"description":"","name":"get_agent"},{"description":"","name":"get_agent_build_status"},{"description":"","name":"get_agent_cost_breakdown"},{"description":"","name":"get_agent_runs"},{"description":"","name":"get_agent_runs_cost"},{"description":"","name":"get_agent_schedule"},{"description":"","name":"get_agent_versions"},{"description":"","name":"get_agents_usage"},{"description":"","name":"get_credit_history"},{"description":"","name":"get_credits_balance"},{"description":"","name":"get_file_download_url"},{"description":"","name":"get_latest_failure"},{"description":"","name":"get_records_summary"},{"description":"","name":"get_run_diagnostics"},{"description":"","name":"get_run_files"},{"description":"","name":"get_run_status"},{"description":"","name":"get_runs_summary"},{"description":"","name":"get_scheduled_runs"},{"description":"","name":"get_space"},{"description":"","name":"get_space_agents"},{"description":"","name":"get_spending_summary"},{"description":"","name":"kill_agent"},{"description":"","name":"list_agent_schedules"},{"description":"","name":"list_agents"},{"description":"","name":"list_spaces"},{"description":"","name":"restore_agent_version"},{"description":"","name":"run_space_agents","risk":{"category":"financial","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can spend money, consume paid credits, or change financially consequential resources. Review findings: Execution consumes metered paid credits."}},{"description":"","name":"search_agents"},{"description":"","name":"search_space_by_name"},{"description":"","name":"start_agent","risk":{"category":"financial","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can spend money, consume paid credits, or change financially consequential resources. Review findings: Execution consumes metered paid credits."}},{"description":"","name":"start_agent_build","risk":{"category":"financial","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can spend money, consume paid credits, or change financially consequential resources. Review findings: Execution consumes metered paid credits."}},{"description":"","name":"stop_agent"},{"description":"","name":"stop_agent_build"},{"description":"","name":"update_agent_schedule","risk":{"category":"code_execution","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can deploy code or change infrastructure and production environments. Review findings: Schedules deploy recurring browser/data workflows that run independently."}}],"tools_count":39,"transport":"streamable-http","type":"web_connector","url":"https://claude.ai/directory/4a77f913-6cde-4b5d-9ab1-f83971992c93","uuid":"dd5c9bf5-821e-53a4-a20b-e8e6bb906d6b","verification_tier":"community","version":""},"query_key":"sequentum sequentum mcp","schema_version":1,"status":"match"}