{"generated_at":"2026-10-14T20:07:22+00:00","item":{"added_at":"2026-09-20","analysis_method":"independent_manual_review_2026-10-10","assessment_status":"provisional","auth_posture":"auth_required","category":"Web Connector","confidence":"medium","description":"Jsonify builds datasets from websites and mobile apps. Review sources, schemas, and preview rows before confirming a pipeline build. Requires an account; pipeline runs show row allowance and maximum charge before starting.","did":"f76a8f49-fc36-47e2-a181-3848a4206042","documentation_url":"https://jsonify.com/connect","editorial_description":{"extended_description":"Query versioned data, inspect schedules or run status, and request pipeline changes. Metering starts with builds and uses row-based pricing; preview rows and test runs are free. Outputs can feed spreadsheets, warehouses, storage, or webhooks.","schema_version":1,"summary":"Jsonify builds datasets from websites and mobile apps.\nReview sources, schemas, and preview rows before confirming a pipeline build.\nRequires an account; pipeline runs show row allowance and maximum charge before starting."},"homepage_url":"","icon":"","id":"f76a8f49-fc36-47e2-a181-3848a4206042","installs":0,"last_scanned":"","license":"","long_description":"Query versioned data, inspect schedules or run status, and request pipeline changes. Metering starts with builds and uses row-based pricing; preview rows and test runs are free. Outputs can feed spreadsheets, warehouses, storage, or webhooks.","manual_assessment":"high","manual_review":{"all_findings":[{"evidence":"The listing says a build agent turns an approved brief into deterministic code that continuously fetches websites/apps; tools build/accept workspaces and run pipelines.","reasoning":"This provisions and executes recurring data-extraction infrastructure.","risk_type":"deployment_infrastructure","severity":"high","title":"Web/mobile extraction pipelines can be built and run","triggering_tools_or_skills":["build_workspace","accept_workspace","run_pipeline","request_work","start_preparation"]},{"evidence":"Jsonify generates deterministic fetching code from a natural-language brief and executes it as a pipeline.","reasoning":"Generated code can access external systems and process data; sandboxing and source restrictions require runtime verification.","risk_type":"code_execution","severity":"high","title":"Generated pipeline code executes against external sources","triggering_tools_or_skills":["build_workspace","run_pipeline","request_work"]},{"evidence":"The inventory exposes query_dataset over produced datasets.","reasoning":"Query scope, language, row limits, and tenant isolation are not shown and must be verified.","risk_type":"query_execution","severity":"high","title":"Datasets can be queried","triggering_tools_or_skills":["query_dataset"]},{"evidence":"Users describe competitor sites, retailers, reviews, listings, or research sources for the pipeline to fetch.","reasoning":"The remote extractor contacts caller-selected destinations; internal-address, authentication, and legal controls are needed.","risk_type":"arbitrary_network_access","severity":"low","title":"User-selected websites and mobile apps are fetched","triggering_tools_or_skills":["request_work","build_workspace","run_pipeline"]},{"evidence":"The pipeline ingests external reviews, listings, papers, and other site/app data.","reasoning":"Scraped content is untrusted and may contain adversarial instructions.","risk_type":"prompt_injection_exposure","severity":"low","title":"Scraped web/app content enters context and datasets","triggering_tools_or_skills":["get_dataset","query_dataset","get_context"]}],"assessment_status":"provisional","auth_and_scopes":"Authentication grants access to Jsonify workspaces, datasets, conversations, pipelines, runs, and build-agent jobs; exact workspace roles are not stated.","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/f76a8f49-fc36-47e2-a181-3848a4206042.json","what_it_proves":"Candidate PR #125 record identity (name \"Jsonify\", publisher, type web_connector) and 15 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/f76a8f49-fc36-47e2-a181-3848a4206042: complete tool_names inventory (15 tools), remote streamable-http endpoint https://factory.jsonify.com/mcp, auth_posture auth_required, directory tier community"}],"primary_reason":"Jsonify generates and runs web/app extraction code and pipelines against user-selected sources.","provisional_risk":null,"publication_recommendation":"accept","requires_runtime_verification":true,"review_key":"claude:f76a8f49-fc36-47e2-a181-3848a4206042","reviewed_at":"2026-10-10","reviewer_notes":"Reviewed all 15 tools; approval is claimed before build, but pipeline sandboxing, credential use, and cancellation semantics need runtime validation.","risk_domains":{"arbitrary_network_access":"true","authz_ambiguity":"false","browser_control":"false","credential_or_admin_actions":"false","deletes_data":"false","deployment_or_infrastructure":"true","dynamic_hidden_actions":"false","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":15,"tools_or_skills_total":15},"name":"Jsonify","needs_runtime_verification":true,"plutonium_url":"https://plutonium.pluto.security/detail.html?planet=claudesec&did=f76a8f49-fc36-47e2-a181-3848a4206042&utm_source=plutonium_analysis_skill&utm_medium=claude_skill&utm_campaign=connector_risk_assessment","privacy_policy_url":"https://jsonify.com/legal","provisional_risk":null,"publication_recommendation":"accept","publisher":"Jsonify","publisher_url":"https://jsonify.com/","remote_url":"https://factory.jsonify.com/mcp","repository_url":"","risk":"high","risk_severity":"high","security_risks":[{"description":"This connector can deploy code or change infrastructure and production environments. Review findings: This provisions and executes recurring data-extraction infrastructure.","evidence":"Web/mobile extraction pipelines can be built and run: The listing says a build agent turns an approved brief into deterministic code that continuously fetches websites/apps; tools build/accept workspaces and run pipelines.","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":"Web/mobile extraction pipelines can be built and run","triggering_tools":["build_workspace","accept_workspace","run_pipeline","request_work","start_preparation"]},{"description":"This connector can execute code, commands, queries, or dynamically selected actions. Review findings: Generated code can access external systems and process data; sandboxing and source restrictions require runtime verification.; Query scope, language, row limits, and tenant isolation are not shown and must be verified.","evidence":"Generated pipeline code executes against external sources: Jsonify generates deterministic fetching code from a natural-language brief and executes it as a pipeline. | Datasets can be queried: The inventory exposes query_dataset over produced datasets.","manual_risk_type":"code_execution,query_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":"Can execute code, queries, or dynamic actions","triggering_tools":["build_workspace","run_pipeline","request_work","query_dataset"]},{"description":"This connector can send caller-provided data to an external service or publisher. Review findings: The remote extractor contacts caller-selected destinations; internal-address, authentication, and legal controls are needed.","evidence":"User-selected websites and mobile apps are fetched: Users describe competitor sites, retailers, reviews, listings, or research sources for the pipeline to fetch.","manual_risk_type":"arbitrary_network_access","remediation":{"block_tool_categories":["sends_externally"],"steps":["Do not include secrets, regulated data, or customer information unless the destination is approved.","Review the recipient's retention and deletion policy before enabling the connector."]},"risk_type":"forwards_data_to_third_party","severity":"low","title":"User-selected websites and mobile apps are fetched","triggering_tools":["request_work","build_workspace","run_pipeline"]}],"signature_status":"unknown","source_code_reviewed":false,"support_url":"paul@jsonify.com","tags":["accesses_messaging","code_deployment","forwards_data_to_third_party","network_access","oauth_required","runs_code","web_connector"],"tools":[{"description":"","name":"accept_workspace","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: This provisions and executes recurring data-extraction infrastructure."}},{"description":"","name":"build_workspace","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: Generated code can access external systems and process data; sandboxing and source restrictions require runtime verification.; Query scope, language, row limits, and tenant isolation are not shown and must be verified."}},{"description":"","name":"cancel_work"},{"description":"","name":"get_context"},{"description":"","name":"get_conversation"},{"description":"","name":"get_dataset"},{"description":"","name":"get_pipeline"},{"description":"","name":"get_run"},{"description":"","name":"list_conversations"},{"description":"","name":"list_datasets"},{"description":"","name":"list_pipelines"},{"description":"","name":"query_dataset","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: Generated code can access external systems and process data; sandboxing and source restrictions require runtime verification.; Query scope, language, row limits, and tenant isolation are not shown and must be verified."}},{"description":"","name":"request_work","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: Generated code can access external systems and process data; sandboxing and source restrictions require runtime verification.; Query scope, language, row limits, and tenant isolation are not shown and must be verified."}},{"description":"","name":"run_pipeline","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: Generated code can access external systems and process data; sandboxing and source restrictions require runtime verification.; Query scope, language, row limits, and tenant isolation are not shown and must be verified."}},{"description":"","name":"start_preparation","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: This provisions and executes recurring data-extraction infrastructure."}}],"tools_count":15,"transport":"streamable-http","type":"web_connector","url":"https://claude.ai/directory/f76a8f49-fc36-47e2-a181-3848a4206042","uuid":"9c72cd63-10ce-54fb-b4da-ae412db43eab","verification_tier":"community","version":""},"query_key":"jsonify jsonify","schema_version":1,"status":"match"}