Skip to content

Arize AX

Official status unknown94 tools reported1 source

Arize AX is a platform for observing, evaluating and improving LLM applications and agents. This server connects your assistant to the Arize AX REST API, so you can inspect and manage your AI app's data from the conversation. What you can do: - Traces and spans: search production traces to debug agent behavior, tool calls and failures. - Datasets: create datasets, add and update examples, and search them. - Experiments: create experiments, log runs, and compare results across prompt or model changes. - Evaluators: build LLM-as-judge evaluators from templates, version them, and run them as tasks on spans or experiments. - Prompts: manage prompts in Prompt Hub, create versions, and set labels like production or staging. - Annotations: define annotation configs, label spans, experiment runs and dataset examples, and manage review queues. - Monitors, projects and spaces: list monitors and browse your workspace. - AI integrations: manage LLM provider credentials for OpenAI, Anthropic, Bedrock, Vertex AI and others. Sign in with your Arize account. An Arize API key also works as a Bearer token.

Connection details

  • Remote endpoint

    Official status unknown

    Transport not established · Authentication not established

    https://api.arize.com/mcp

    Use the source documentation for installation instructions. No command has been inferred.

    Connection source and version

    Reported by Claude logoClaude connector directory

94 tool names reported

Source-declared tools. Availability can depend on the connection and permissions; these are not runtime test results.

Claude connector directory ↗ · Observed 2026-11-01 · Completeness not established

  • annotate_annotation_queue_record
  • annotate_dataset_examples
  • annotate_experiment_runs
  • annotate_spans
  • assign_annotation_queue_record
  • cancel_task_run
  • create_annotation_config
  • create_annotation_queue
  • create_annotation_queue_record
  • create_dataset
  • create_evaluator
  • create_evaluator_version
  • create_experiment
  • create_integration
  • create_project
  • create_prompt
  • create_prompt_version
  • create_space
  • create_task
  • delete_annotation_config
  • delete_annotation_queue
  • delete_annotation_queue_record
  • delete_dataset
  • delete_dataset_examples
  • delete_evaluator
  • delete_evaluator_versions
  • delete_experiment
  • delete_integration
  • delete_project
  • delete_prompt
  • delete_prompt_version_label
  • delete_space
  • delete_spans
  • delete_task
  • get_annotation_config
  • get_annotation_queue
  • get_dataset
  • get_evaluator
  • get_evaluator_version
  • get_experiment
  • get_integration
  • get_monitor
  • get_organization
  • get_project
  • get_prompt
  • get_prompt_label
  • get_prompt_version
  • get_space
  • get_task
  • get_task_run
  • insert_dataset_examples
  • insert_experiment_runs
  • list_annotation_config_tags
  • list_annotation_configs
  • list_annotation_queue_records
  • list_annotation_queue_tags
  • list_annotation_queues
  • list_dataset_tags
  • list_datasets
  • list_evaluator_tags
  • list_evaluator_templates
  • list_evaluator_versions
  • list_evaluators
  • list_experiment_tags
  • list_experiments
  • list_integrations
  • list_monitor_tags
  • list_monitors
  • list_organizations
  • list_project_tags
  • list_projects
  • list_prompt_tags
  • list_prompt_versions
  • list_prompts
  • list_spaces
  • list_spans
  • list_task_runs
  • list_tasks
  • list_traces
  • search_dataset_examples
  • search_experiment_runs
  • server_info
  • set_prompt_version_label
  • trigger_task_run
  • update_annotation_config
  • update_annotation_queue
  • update_dataset
  • update_dataset_examples
  • update_evaluator
  • update_integration
  • update_project
  • update_prompt
  • update_space
  • update_task

Sources and provenance

Grouping basis: endpoint. Exact endpoints or package sets can be grouped automatically; other relationships require evidence review. This does not verify publisher affiliation, connection health, or equivalent capabilities.

Search API Tracker

Curated starting points