# Verial

## Docs

- [Guided Onboarding](https://docs.verial.ai/ai-agents/onboarding.md): Point a coding agent at Verial and have it wire up a first evaluation end to end.
- [Verial for AI Agents](https://docs.verial.ai/ai-agents/overview.md): Give AI coding agents the tools to build, author, and drive benchmarks on Verial.
- [Agent Skills](https://docs.verial.ai/ai-agents/skills.md): Reusable expertise for AI coding agents operating on Verial.
- [Authentication](https://docs.verial.ai/api-reference/authentication.md): Authenticate requests to the Verial internal API and the public v1 API.
- [Create benchmark](https://docs.verial.ai/api-reference/benchmarks/create.md): Create a benchmark of tasks against an environment.
- [Delete benchmark](https://docs.verial.ai/api-reference/benchmarks/delete.md): Delete a benchmark.
- [Get benchmark](https://docs.verial.ai/api-reference/benchmarks/get.md): Retrieve a benchmark by ID.
- [List benchmarks](https://docs.verial.ai/api-reference/benchmarks/list.md): Retrieve benchmarks in your organization.
- [Update benchmark](https://docs.verial.ai/api-reference/benchmarks/update.md): Update benchmark metadata or tasks.
- [Submit CDS feedback](https://docs.verial.ai/api-reference/cds/feedback.md): Submit feedback for a CDS Hook response.
- [Invoke CDS hook](https://docs.verial.ai/api-reference/cds/invoke.md): Trigger a CDS Hook invocation.
- [List CDS services](https://docs.verial.ai/api-reference/cds/list.md): List registered CDS Hooks services.
- [Register CDS service](https://docs.verial.ai/api-reference/cds/register.md): Register a CDS Hooks service.
- [Check eligibility](https://docs.verial.ai/api-reference/clearinghouse/check-eligibility.md): Perform an eligibility check via a clearinghouse.
- [Create dataset](https://docs.verial.ai/api-reference/datasets/create.md): Create a new dataset.
- [Delete dataset](https://docs.verial.ai/api-reference/datasets/delete.md): Delete a dataset and its files.
- [Get dataset](https://docs.verial.ai/api-reference/datasets/get.md): Retrieve details for a specific dataset.
- [List datasets](https://docs.verial.ai/api-reference/datasets/list.md): Retrieve datasets in your organization.
- [Update dataset](https://docs.verial.ai/api-reference/datasets/update.md): Update dataset metadata or configuration.
- [Create environment](https://docs.verial.ai/api-reference/environments/create.md): Create a reusable simulated health system definition.
- [Delete environment](https://docs.verial.ai/api-reference/environments/delete.md): Delete an environment definition.
- [Get environment](https://docs.verial.ai/api-reference/environments/get.md): Retrieve details for a specific environment.
- [List environments](https://docs.verial.ai/api-reference/environments/list.md): Retrieve environments in your organization.
- [Update environment](https://docs.verial.ai/api-reference/environments/update.md): Update environment configuration.
- [Errors](https://docs.verial.ai/api-reference/errors.md): Error response format and common error codes.
- [Get fax](https://docs.verial.ai/api-reference/fax/get.md): Retrieve a submitted fax.
- [Send fax](https://docs.verial.ai/api-reference/fax/send.md): Submit a fax to a simulated fax service.
- [SMART configuration](https://docs.verial.ai/api-reference/fhir/smart-config.md): Fetch the SMART-on-FHIR configuration for a service.
- [FHIR token](https://docs.verial.ai/api-reference/fhir/token.md): Issue an OAuth token for FHIR access.
- [Send HL7 message](https://docs.verial.ai/api-reference/hl7/send.md): Send an HL7 v2 message to a simulator.
- [Introduction](https://docs.verial.ai/api-reference/introduction.md): The Verial API. Healthcare simulated environment platform for defining environments, publishing benchmarks, and running scored rollouts.
- [Get message](https://docs.verial.ai/api-reference/messages/get.md): Retrieve a specific message.
- [List messages](https://docs.verial.ai/api-reference/messages/list.md): Retrieve the message history for a service.
- [Send message](https://docs.verial.ai/api-reference/messages/send.md): Send a chat message to a simulator.
- [Pagination](https://docs.verial.ai/api-reference/pagination.md): How cursor-based pagination works in the Verial API.
- [Get prior auth](https://docs.verial.ai/api-reference/payer/get-prior-auth.md): Retrieve a prior authorization by auth number.
- [List payer patients](https://docs.verial.ai/api-reference/payer/list-patients.md): List patients covered by a payer simulator.
- [Submit prior auth](https://docs.verial.ai/api-reference/payer/submit-prior-auth.md): Submit a prior authorization request.
- [Benchmarks](https://docs.verial.ai/api-reference/resources/benchmarks.md): Groups of tasks with criteria.
- [Criteria](https://docs.verial.ai/api-reference/resources/criteria.md): Typed assertions that the verification engine runs against a task run.
- [Criterion Runs](https://docs.verial.ai/api-reference/resources/criterion-runs.md): Per-criterion results produced by the verification engine.
- [Datasets](https://docs.verial.ai/api-reference/resources/datasets.md): Synthetic patient data for populating sandboxes.
- [Environments](https://docs.verial.ai/api-reference/resources/environments.md): Simulated health systems composed of linked simulators.
- [Overview](https://docs.verial.ai/api-reference/resources/overview.md): Resources available in the Verial API.
- [Playgrounds](https://docs.verial.ai/api-reference/resources/playgrounds.md): Running environment instances with live sandboxes.
- [Benchmark Runs](https://docs.verial.ai/api-reference/resources/runs.md): Benchmark executions with scores and verdicts.
- [Sandboxes](https://docs.verial.ai/api-reference/resources/sandboxes.md): Running simulator instances with credentials and event logs.
- [Simulators](https://docs.verial.ai/api-reference/resources/simulators.md): Reusable simulator definitions for healthcare interfaces.
- [Task Runs](https://docs.verial.ai/api-reference/resources/task-runs.md): Task-level results within a benchmark run.
- [Tasks](https://docs.verial.ai/api-reference/resources/tasks.md): Individual test cases within benchmarks.
- [Create run](https://docs.verial.ai/api-reference/runs/create.md): Execute a benchmark and produce a run.
- [Get run](https://docs.verial.ai/api-reference/runs/get.md): Retrieve details for a specific run.
- [List runs](https://docs.verial.ai/api-reference/runs/list.md): Retrieve runs in your organization.
- [Update run](https://docs.verial.ai/api-reference/runs/update.md): Update run metadata.
- [verial auth](https://docs.verial.ai/cli/auth.md): Manage CLI authentication with your organization API key.
- [verial benchmark-runs](https://docs.verial.ai/cli/benchmark-runs.md): Start, inspect, and wait on benchmark runs from the command line.
- [verial benchmarks](https://docs.verial.ai/cli/benchmarks.md): Create and manage benchmarks from the command line.
- [verial config](https://docs.verial.ai/cli/config.md): View and update persistent CLI configuration.
- [verial datasets](https://docs.verial.ai/cli/datasets.md): Manage datasets that back simulators in your environments.
- [verial environments](https://docs.verial.ai/cli/environments.md): Create, inspect, and instantiate environments from the command line.
- [Installation](https://docs.verial.ai/cli/installation.md): Install the Verial CLI via npm or npx.
- [Overview](https://docs.verial.ai/cli/overview.md): Command-line interface for running benchmarks and managing resources.
- [verial playgrounds](https://docs.verial.ai/cli/playgrounds.md): Instantiate and tear down playgrounds from the command line.
- [verial sandboxes](https://docs.verial.ai/cli/sandboxes.md): Provision, inspect, and tear down sandboxes from the command line.
- [verial simulators](https://docs.verial.ai/cli/simulators.md): Create, inspect, and manage simulators from the command line.
- [verial task-runs](https://docs.verial.ai/cli/task-runs.md): Inspect individual task runs within a benchmark run.
- [Authoring a Benchmark](https://docs.verial.ai/guides/authoring-a-benchmark.md): Compose an environment, build a benchmark of tasks with typed criteria, and publish a versioned slug.
- [Benchmarks](https://docs.verial.ai/guides/concepts/benchmarks.md): Collections of tasks that test specific healthcare AI agent capabilities.
- [Criteria](https://docs.verial.ai/guides/concepts/criteria.md): Typed assertions attached to a task. The verification engine runs them after a rollout to score the task.
- [Datasets](https://docs.verial.ai/guides/concepts/datasets.md): Synthetic data loaded into simulator sandboxes at rollout time.
- [Environments](https://docs.verial.ai/guides/concepts/environments.md): Simulated health systems that your agent connects to during benchmarks.
- [Interactions](https://docs.verial.ai/guides/concepts/interactions.md): Recorded evidence captured on each sandbox during a rollout.
- [Playgrounds](https://docs.verial.ai/guides/concepts/playgrounds.md): A provisioned, running environment ready for agent rollouts.
- [Benchmark Runs](https://docs.verial.ai/guides/concepts/runs.md): Rollouts of a benchmark, scored by the verification engine.
- [Sandboxes](https://docs.verial.ai/guides/concepts/sandboxes.md): A running simulator instance with credentials, events, and branchable datasets.
- [Tasks](https://docs.verial.ai/guides/concepts/tasks.md): One test case within a benchmark. Drives a single rollout and is scored by its criteria.
- [Verification](https://docs.verial.ai/guides/concepts/verification.md): How the verification engine scores a task run by running criteria against recorded sandbox state.
- [Run a Benchmark in GitHub Actions](https://docs.verial.ai/guides/github-actions.md): Run a Verial benchmark on every PR. Fail the check if the score drops.
- [Welcome](https://docs.verial.ai/guides/index.md): Verial is the benchmarking platform for healthcare AI agents.
- [Interactions](https://docs.verial.ai/guides/observability/interactions.md): Raw evidence captured per sandbox during a rollout.
- [Observability Overview](https://docs.verial.ai/guides/observability/overview.md): What Verial captures during a rollout, and how to read it back.
- [Run Results](https://docs.verial.ai/guides/observability/run-results.md): Read a completed benchmark run top-down: verdict, axes, per-criterion evidence.
- [Quick Start](https://docs.verial.ai/guides/quickstart.md): Run a published benchmark against your agent in 10 minutes using the v1 API.
- [Web Simulator Quickstart](https://docs.verial.ai/guides/quickstart-web-sim.md): Spin up a RadMD web portal sandbox and log in from a browser.
- [Running a Benchmark](https://docs.verial.ai/guides/running-a-benchmark.md): Browse published benchmarks, drive your agent through a rollout, and compare runs over time.
- [Scheduled Runs](https://docs.verial.ai/guides/scheduled-runs.md): Re-run a benchmark on a cadence to catch regressions from upstream model or tool changes.
- [CDS Hooks](https://docs.verial.ai/guides/simulators/cds-hooks.md): Clinical Decision Support Hooks registration, invocation, and feedback.
- [Fax](https://docs.verial.ai/guides/simulators/fax.md): Simulated fax numbers for inbound and outbound documents with OCR evidence.
- [FHIR](https://docs.verial.ai/guides/simulators/fhir.md): A fully functional FHIR R4 EHR backed by GCP Healthcare API, with SMART on FHIR auth.
- [Files (SFTP)](https://docs.verial.ai/guides/simulators/files-sftp.md): SFTP-style file drops for batch intake and outbound deliveries.
- [HL7](https://docs.verial.ai/guides/simulators/hl7.md): HL7v2 messaging for ADT, ORU, ORM, and SIU feeds.
- [Messages](https://docs.verial.ai/guides/simulators/messages.md): Conversational messaging with patient personas or portal chat agents.
- [Simulators](https://docs.verial.ai/guides/simulators/overview.md): Simulated healthcare interfaces your agent connects to.
- [Voice](https://docs.verial.ai/guides/simulators/voice.md): Simulated phone / IVR lines with recorded call transcripts.
- [Web Portal](https://docs.verial.ai/guides/simulators/web-portal.md): Browser-accessible simulations of real payer web portals with a matching REST API.
- [X12](https://docs.verial.ai/guides/simulators/x12.md): Clearinghouse-style X12 EDI transactions for eligibility, claims, and prior auth.
- [Solver Keys](https://docs.verial.ai/guides/solver-keys.md): Create and rotate the Bearer tokens that drive benchmark runs via the public v1 API.
- [Webhooks](https://docs.verial.ai/guides/webhooks.md): Subscribe an HTTPS endpoint to events from Verial so your systems react when runs finish.
- [Best Practices](https://docs.verial.ai/mcp/best-practices.md): Tips for getting the best results from the Verial MCP.
- [MCP Server](https://docs.verial.ai/mcp/overview.md): Give AI agents direct access to Verial's simulation platform.
- [Setup](https://docs.verial.ai/mcp/setup.md): Connect the Verial MCP server to your AI agent.
- [Benchmark Runs](https://docs.verial.ai/mcp/tools/benchmark-runs.md): Manage benchmark runs.
- [Benchmarks](https://docs.verial.ai/mcp/tools/benchmarks.md): Manage benchmarks (groups of tasks with criteria).
- [Criteria](https://docs.verial.ai/mcp/tools/criteria.md): Manage task criteria (typed assertions).
- [Criterion Runs](https://docs.verial.ai/mcp/tools/criterion-runs.md): View per-criterion results within a task run.
- [Datasets](https://docs.verial.ai/mcp/tools/datasets.md): Manage datasets (FHIR bundles, file sets).
- [Environments](https://docs.verial.ai/mcp/tools/environments.md): Manage simulated healthcare environments.
- [Playgrounds](https://docs.verial.ai/mcp/tools/playgrounds.md): Manage simulated healthcare playgrounds.
- [Sandboxes](https://docs.verial.ai/mcp/tools/sandboxes.md): Manage sandbox instances (running simulators).
- [Simulators](https://docs.verial.ai/mcp/tools/simulators.md): Manage simulators (FHIR, HL7, voice, fax, etc.).
- [Task Runs](https://docs.verial.ai/mcp/tools/task-runs.md): Manage task runs within a benchmark run.
- [Tasks](https://docs.verial.ai/mcp/tools/tasks.md): Manage benchmark tasks.
- [Eligibility Check](https://docs.verial.ai/mcp/workflows/eligibility-check.md): Verify insurance eligibility through a clearinghouse simulator.
- [Environment Setup](https://docs.verial.ai/mcp/workflows/environment-setup.md): Create a complete simulation environment from scratch.
- [Prior Auth Testing](https://docs.verial.ai/mcp/workflows/prior-auth.md): Define and run a prior authorization benchmark.
- [Errors](https://docs.verial.ai/sdk/errors.md): Typed error classes, retry behavior, and status code mapping.
- [Overview](https://docs.verial.ai/sdk/index.md): TypeScript SDK for Verial.
- [Installation](https://docs.verial.ai/sdk/installation.md): Install and configure the Verial TypeScript SDK.
- [Pagination](https://docs.verial.ai/sdk/pagination.md): Cursor-based iteration across SDK list endpoints.
- [Benchmarks](https://docs.verial.ai/sdk/resources/benchmarks.md): SDK methods for the Benchmarks resource.
- [Datasets](https://docs.verial.ai/sdk/resources/datasets.md): SDK methods for the Datasets resource.
- [Environments](https://docs.verial.ai/sdk/resources/environments.md): SDK methods for the Environments resource.
- [Playgrounds](https://docs.verial.ai/sdk/resources/playgrounds.md): SDK methods for the Playgrounds resource.
- [Runs](https://docs.verial.ai/sdk/resources/runs.md): SDK methods for the Benchmark Runs resource.
- [Sandboxes](https://docs.verial.ai/sdk/resources/sandboxes.md): SDK methods for the Sandboxes resource.
- [Simulators](https://docs.verial.ai/sdk/resources/simulators.md): SDK methods for the Simulators resource.
- [Task Runs](https://docs.verial.ai/sdk/resources/task-runs.md): SDK methods for the Task Runs resource.
- [Tasks](https://docs.verial.ai/sdk/resources/tasks.md): SDK methods for the Tasks resource.
- [Usage](https://docs.verial.ai/sdk/usage.md): Common workflows with the Verial SDK.

## OpenAPI Specs

- [openapi](https://docs.verial.ai/openapi.json)