{"generated_at":"2026-10-14T20:07:22+00:00","item":{"added_at":"2026-08-06","analysis_method":"independent_manual_review_2026-10-10","assessment_status":"provisional","auth_posture":"no_auth","category":"Web Connector","confidence":"medium","description":"Mixer4AI connects participating business AI agents through a directory and intent exchange. Publish capability profiles or business needs and search for matching agents. Registration tools support business-attestation checks and agent approval.","did":"eac4518e-ced6-417e-8407-e55553c7cbe0","documentation_url":"https://mixer4ai.com/docs","editorial_description":{"extended_description":"Intents can describe requests for proposals or problem statements. Inspect agent cards, track self-registration, and check domain or endpoint challenges before completing registration.","schema_version":1,"summary":"Mixer4AI connects participating business AI agents through a directory and intent exchange.\nPublish capability profiles or business needs and search for matching agents.\nRegistration tools support business-attestation checks and agent approval."},"homepage_url":"","icon":"","id":"eac4518e-ced6-417e-8407-e55553c7cbe0","installs":0,"last_scanned":"","license":"","long_description":"Intents can describe requests for proposals or problem statements. Inspect agent cards, track self-registration, and check domain or endpoint challenges before completing registration.","manual_assessment":"high","manual_review":{"all_findings":[{"evidence":"The no-auth inventory exposes agent-card self-registration, domain and endpoint challenge checks, and a final approval action.","reasoning":"These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.","risk_type":"credential_admin","severity":"high","title":"Unauthenticated self-registration can approve an agent identity","triggering_tools_or_skills":["self_register_agent_card","check_self_registration_domain_challenge","check_self_registration_endpoint_challenge","approve_agent_self_registration"]},{"evidence":"The listing says participants publish needs, RFPs, and problem statements into a searchable exchange, and publish_intent is the committing tool.","reasoning":"Publishing makes user-authored business demand visible to other agents and businesses as the registered participant.","risk_type":"messaging_as_user","severity":"high","title":"Structured intents can be published to the exchange","triggering_tools_or_skills":["publish_intent"]},{"evidence":"The tools register agent capability cards and update self-registration state in the public directory.","reasoning":"They persist identity and capability metadata that other participants may rely on for trust decisions.","risk_type":"data_modification","severity":"medium","title":"Agent cards and registration records can be created or updated","triggering_tools_or_skills":["register_agent_card","self_register_agent_card","update_self_registration"]},{"evidence":"The packet marks the server no_auth while exposing public registration, approval, update, and publishing actions; only challenge tool names indicate protection.","reasoning":"Runtime schemas and challenge binding must be checked to ensure one caller cannot approve, update, or publish for another identity.","risk_type":"authz_ambiguity","severity":"low","title":"No-auth writes depend on undocumented challenges","triggering_tools_or_skills":["approve_agent_self_registration","publish_intent","register_agent_card"]}],"assessment_status":"provisional","auth_and_scopes":"No sign-in is declared; self-registration relies on domain and endpoint challenge tools, but challenge binding and authorization rules are not in the packet.","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/eac4518e-ced6-417e-8407-e55553c7cbe0.json","what_it_proves":"Candidate PR #125 record identity (name \"Mixer4AI\", publisher, type web_connector) and 12 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/eac4518e-ced6-417e-8407-e55553c7cbe0: complete tool_names inventory (12 tools), remote streamable-http endpoint https://mcp.mixer4ai.com/, auth_posture no_auth, directory tier community"}],"primary_reason":"Mixer4AI exposes no-auth identity registration/approval and public intent publishing guarded only by undocumented challenge semantics.","provisional_risk":null,"publication_recommendation":"accept","requires_runtime_verification":true,"review_key":"claude:eac4518e-ced6-417e-8407-e55553c7cbe0","reviewed_at":"2026-10-10","reviewer_notes":"Reviewed all 12 tools; the challenge workflow is plausible but source code and parameter schemas are unavailable, so runtime authorization testing is essential.","risk_domains":{"arbitrary_network_access":"false","authz_ambiguity":"true","browser_control":"false","credential_or_admin_actions":"true","deletes_data":"false","deployment_or_infrastructure":"false","dynamic_hidden_actions":"false","filesystem_or_repo_access":"false","forwards_data_to_third_party":"true","modifies_data":"true","persistent_credentials":"false","prompt_injection_exposure":"true","provenance_uncertainty":"true","reads_private_data":"false","runs_code_or_queries":"false","sends_messages_as_user":"true","spends_or_moves_money":"false"},"tools_or_skills_reviewed":12,"tools_or_skills_total":12},"name":"Mixer4AI","needs_runtime_verification":true,"plutonium_url":"https://plutonium.pluto.security/detail.html?planet=claudesec&did=eac4518e-ced6-417e-8407-e55553c7cbe0&utm_source=plutonium_analysis_skill&utm_medium=claude_skill&utm_campaign=connector_risk_assessment","privacy_policy_url":"https://mixer4ai.com/privacy","provisional_risk":null,"publication_recommendation":"accept","publisher":"Locus Consulting, LLC","publisher_url":"https://locusconsulting.com/","remote_url":"https://mcp.mixer4ai.com/","repository_url":"","risk":"high","risk_severity":"high","security_risks":[{"description":"This connector can create, update, configure, or otherwise change connected resources. Review findings: These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.; They persist identity and capability metadata that other participants may rely on for trust decisions.","evidence":"Unauthenticated self-registration can approve an agent identity: The no-auth inventory exposes agent-card self-registration, domain and endpoint challenge checks, and a final approval action. | Agent cards and registration records can be created or updated: The tools register agent capability cards and update self-registration state in the public directory.","manual_risk_type":"credential_admin,data_modification","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":"Can change connected resources","triggering_tools":["self_register_agent_card","check_self_registration_domain_challenge","check_self_registration_endpoint_challenge","approve_agent_self_registration","register_agent_card","update_self_registration"]},{"description":"This connector can publish or send messages and content using the connected identity. Review findings: Publishing makes user-authored business demand visible to other agents and businesses as the registered participant.","evidence":"Structured intents can be published to the exchange: The listing says participants publish needs, RFPs, and problem statements into a searchable exchange, and publish_intent is the committing tool.","manual_risk_type":"messaging_as_user","remediation":{"block_tool_categories":["sends_externally"],"steps":["Require review before messages, posts, or publications are sent.","Restrict delivery to test or internal recipients where possible."]},"risk_type":"sends_messages_as_you","severity":"high","title":"Structured intents can be published to the exchange","triggering_tools":["publish_intent"]}],"signature_status":"unknown","source_code_reviewed":false,"support_url":"info@locusconsulting.com","tags":["modifies_data","network_access","sends_messages_as_you","web_connector"],"tools":[{"description":"","name":"approve_agent_self_registration","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: These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.; They persist identity and capability metadata that other participants may rely on for trust decisions."}},{"description":"","name":"check_self_registration_domain_challenge","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: These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.; They persist identity and capability metadata that other participants may rely on for trust decisions."}},{"description":"","name":"check_self_registration_endpoint_challenge","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: These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.; They persist identity and capability metadata that other participants may rely on for trust decisions."}},{"description":"","name":"find_verified_agent"},{"description":"","name":"get_agent_card"},{"description":"","name":"get_self_registration_status"},{"description":"","name":"publish_intent","risk":{"category":"sends_externally","level":"high","recommendation":"Follow the connector-level remediation and require approval for consequential actions.","why":"This connector can publish or send messages and content using the connected identity. Review findings: Publishing makes user-authored business demand visible to other agents and businesses as the registered participant."}},{"description":"","name":"register_agent_card","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: These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.; They persist identity and capability metadata that other participants may rely on for trust decisions."}},{"description":"","name":"search_directory"},{"description":"","name":"search_published_intents"},{"description":"","name":"self_register_agent_card","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: These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.; They persist identity and capability metadata that other participants may rely on for trust decisions."}},{"description":"","name":"update_self_registration","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: These operations establish and attest an external business-agent identity; challenge integrity is security-critical and cannot be inferred from names alone.; They persist identity and capability metadata that other participants may rely on for trust decisions."}}],"tools_count":12,"transport":"streamable-http","type":"web_connector","url":"https://claude.ai/directory/eac4518e-ced6-417e-8407-e55553c7cbe0","uuid":"1b3fb312-5e59-5ad8-af17-e8832ba244eb","verification_tier":"community","version":""},"query_key":"mixer4ai","schema_version":1,"status":"match"}