generated: '2026-08-14'
method: searched
status: published
source: https://developer.regal.ai/docs/regal-mcp
docs: https://developer.regal.ai/docs/regal-mcp
sso_docs: https://developer.regal.ai/docs/okta-sso-for-mcp
announcement: https://www.regal.ai/blog/regal-mcp
deployment:
  mode: remote
  endpoint: https://mcp.regal.ai/v1/external-mcp/mcp
  auth: oauth
  verified: probed
  note: 'Regal ships a hosted remote MCP server. It is reachable right now — an unauthenticated POST of
    tools/list to the endpoint returns HTTP 401 with a WWW-Authenticate: Bearer error="invalid_token"
    challenge and a resource_metadata pointer, which is the correct MCP OAuth handshake, not an outage.
    Regal also documents `npx mcp-remote <endpoint>` for Claude Desktop, but that is the generic mcp-remote
    bridge to this same hosted URL, not a first-party stdio package — the mode is remote, not both.'
  probe_prior: (never probed)
  probe: gated
  probe_why: RFC 9728 challenge on the MCP path only
  checked: '2026-09-11'
  source: claimed-backlog re-probe 2026-09-11
server:
  name: RegalMCP
  transport: http
  url: https://mcp.regal.ai/v1/external-mcp/mcp
  protocol: Model Context Protocol
  tenancy: All tools are scoped to the authenticated brand; Regal states no cross-tenant access is possible.
authentication:
  type: oauth2
  provider: Okta (same login as the Regal application)
  flows:
  - authorization_code
  - refresh_token
  pkce: S256
  dynamic_client_registration: true
  registration_endpoint: https://mcp.regal.ai/v1/external-mcp/register
  authorization_endpoint: https://mcp.regal.ai/v1/external-mcp/authorize
  token_endpoint: https://mcp.regal.ai/v1/external-mcp/token
  revocation_endpoint: https://mcp.regal.ai/v1/external-mcp/revoke
  scopes:
  - openid
  - profile
  - email
  - offline_access
  metadata:
  - well-known/regal-ai-oauth-authorization-server.json
  - well-known/regal-ai-openid-configuration.json
  - well-known/regal-ai-oauth-protected-resource.json
  note: Organizations on Okta SSO need one-time admin configuration before MCP authentication works (https://developer.regal.ai/docs/okta-sso-for-mcp).
clients_documented:
- client: Cursor
  method: mcp.json url entry
- client: Claude Desktop
  method: connectors UI or npx mcp-remote config
- client: Claude Code
  method: claude mcp add --transport http --scope user RegalMCP <endpoint>
- client: Codex
  method: ~/.codex/config.toml [mcp_servers.RegalMCP] url
- client: ChatGPT
  method: Connectors (Business/Enterprise workspace
  Developer Mode): null
skills:
  published: true
  note: Regal pairs its MCP tools with "Skills" — packaged workflow instructions the server exposes so
    external clients get guided workflows rather than raw tool access. The `list-regal-skills` tool enumerates
    them; the skill bodies are not published publicly.
tools_source: Verbatim from the published tool tables at https://developer.regal.ai/docs/regal-mcp (fetched
  2026-08-14). Live tools/list introspection is OAuth-gated (HTTP 401 invalid_token), so per-tool inputSchema
  is NOT captured here — obtaining it requires an authenticated Regal session. Names and descriptions
  below are the provider's own.
tool_counts:
  read: 44
  write: 16
  total: 60
tools:
- name: get-brand
  mode: read
  description: Fetch your brand's configuration, including telephony, compliance, and queue settings.
- name: query-regal-kb
  mode: read
  description: Query the Regal knowledge base with a natural language question.
- name: list-regal-skills
  mode: read
  description: List available skills.
- name: list-ai-agents
  mode: read
  description: List AI agents configured for your brand, with pagination and channel-type filtering.
  aliases:
  - list-agents
- name: fetch-agent
  mode: read
  description: Fetch an active version of a specific agent variant, including its full configuration and
    prompt.
- name: fetch-latest-agent
  mode: read
  description: Fetch highest version (active, draft or deactivated) of a specific agent variant; use before
    edit-agent for updates.version.
- name: fetch-agent-version
  mode: read
  description: Fetch a specific version of an agent variant (including historical ones).
- name: list-ai-agent-variants
  mode: read
  description: List all variants for a given agent, including traffic allocation.
- name: list-ai-agent-versions
  mode: read
  description: List all versions for a given agent variant.
- name: fetch-test-cases
  mode: read
  description: Fetch test cases associated with an agent.
- name: explore-agent-context
  mode: read
  description: Returns workflow instructions for building and understanding AI agent configurations.
- name: summarize-agent-plan
  mode: read
  description: Returns a plan template to guide agent design decisions.
- name: generate-agent-config
  mode: read
  description: Returns section-by-section guidance for editing an agent config.
- name: get-prompting-best-practices
  mode: read
  description: Returns Regal's static prompting reference for agent builders.
- name: get-ai-agent-tool-context
  mode: read
  description: Returns the tool-context reference used when building agent tool configurations.
- name: fetch-simulation-transcript
  mode: read
  description: Fetch the transcript from a completed simulation run.
- name: list-recordings
  mode: read
  description: List call recordings, with filters for date range, agent, disposition, and phone number.
- name: fetch-recording-transcripts
  mode: read
  description: Fetch transcripts for one or more recordings (up to 5 SIDs per call).
- name: get-trackers-lookup
  mode: read
  description: Return the list of tracker IDs and names defined for your brand.
- name: list-dispositions
  mode: read
  description: Lists dispositions in the account.
- name: list-analysis-data-points
  mode: read
  description: List analysis data point definitions used for call scoring and QA.
- name: list-journeys
  mode: read
  description: List journeys with search and pagination support.
- name: fetch-journey
  mode: read
  description: Fetch a full journey definition, including its flow and node configuration. Webhook auth
    headers are redacted.
- name: list-journey-versions
  mode: read
  description: List all saved versions of a journey.
- name: fetch-journey-version
  mode: read
  description: Fetch a specific historical version of a journey for diffing or review.
- name: list-profile-journey-executions
  mode: read
  description: List journey executions for a given contact profile.
- name: get-profile-journey-execution-nodes
  mode: read
  description: Get the node-level detail of a specific journey execution.
- name: lookup-profile
  mode: read
  description: Look up a contact profile by phone, email, name, or external ID. Returns multiple candidates
    if the query is ambiguous.
- name: fetch-profile
  mode: read
  description: Fetch a full contact profile by profile ID.
- name: fetch-profile-events
  mode: read
  description: Fetch event history for a contact profile, with type and cursor-based pagination.
- name: list-profile-versions
  mode: read
  description: List all saved versions of a contact profile.
- name: fetch-profile-version
  mode: read
  description: Fetch a specific historical snapshot of a profile.
- name: list-users
  mode: read
  description: List agent and user accounts in your brand.
- name: fetch-user
  mode: read
  description: Fetch a specific user account by email, including queue assignments.
- name: list-queues
  mode: read
  description: List routing queues and their eligibility rules.
- name: list-tasks
  mode: read
  description: List tasks by state, queue, type, or phone number, with pagination.
- name: fetch-task
  mode: read
  description: Fetch a specific task by SID.
- name: get-task-events
  mode: read
  description: Fetch routing event history for a task.
- name: get-routing-rules
  mode: read
  description: Get routing rule definitions for your brand.
  known_gap: requires a UI session; may error on the external MCP endpoint
- name: get-routing-rule-versions
  mode: read
  description: Get version history for a named routing rule.
  known_gap: requires a UI session; may error on the external MCP endpoint
- name: get-simulation-progress
  mode: read
  description: Poll the progress of a running simulation.
- name: list-business-profiles
  mode: read
  description: List your business profile(s).
- name: list-active-phone-numbers
  mode: read
  description: Lists your Regal provisioned phone numbers.
- name: list-brandedphone-numbers
  mode: read
  description: Lists your phone numbers that were uploaded for branded caller id and/or spam remediation.
- name: create-agent
  mode: write
  description: Create a new AI agent.
- name: edit-agent
  mode: write
  description: Edit an existing agent (saves as a draft version).
- name: create-ai-agent-variant
  mode: write
  description: Create a new variant for an existing agent.
- name: set-variant-traffic-allocations
  mode: write
  description: Set traffic allocation percentages across all variants of an agent.
- name: create-test-cases
  mode: write
  description: Create test cases for an agent.
- name: edit-test-case
  mode: write
  description: Edit an existing test case.
- name: delete-test-cases
  mode: write
  description: Delete one or more test cases.
- name: start-simulation-run
  mode: write
  description: Start a simulation run against an agent and a set of test cases.
- name: create-dispositions
  mode: write
  description: Creates dispositions.
- name: edit-dispositions
  mode: write
  description: Updates existing dispositions.
- name: create-analysis-data-point
  mode: write
  description: Creates a new custom AI analysis data point.
- name: edit-analysis-data-points
  mode: write
  description: Edit one or more existing custom AI analysis data points.
- name: delete-analysis-data-points
  mode: write
  description: Delete one or more existing custom AI analysis data points.
- name: create-branded-phone-number
  mode: write
  description: Adds a phone number for branded caller id and/or spam remediation.
- name: update-branded-phone-number
  mode: write
  description: Updates a phone number for branded caller id and/or spam remediation, including registering/unregistering
    numbers for carriers.
- name: delete-branded-phone-number
  mode: write
  description: Deletes a phone number from branded caller id and/or spam remediation.
governance:
  human_approval: Regal states write tools require explicit human approval at the client level before
    execution, and that current write tools only save drafts — they do not publish or go live without
    a separate action.
  known_gaps:
  - get-routing-rules and get-routing-rule-versions currently require a UI session and may return an error
    from the external MCP endpoint; Regal documents list-queues as the workaround.
x-evidence:
- fetched: '2026-08-14'
  url: https://developer.regal.ai/docs/regal-mcp.md
  http_status: 200
- fetched: '2026-08-14'
  url: https://mcp.regal.ai/v1/external-mcp/mcp
  method: POST tools/list
  http_status: 401
  detail: 'WWW-Authenticate: Bearer error="invalid_token"'
- fetched: '2026-08-14'
  url: https://mcp.regal.ai/.well-known/oauth-authorization-server
  http_status: 200
- fetched: '2026-08-14'
  url: https://mcp.regal.ai/.well-known/oauth-protected-resource/v1/external-mcp/mcp
  http_status: 200