Free cookie consent management tool by TermsFeed

Appian MCP Server

On This Page
The capabilities described on this page are included in Appian's advanced and premium capability tiers. Usage limits may apply.

Appian MCP Server lets system administrators, developers, and end users connect external AI applications to Appian through Model Context Protocol (MCP), an open industry standard for connecting AI applications to external tools and data sources.

When Appian MCP Server is enabled, users can connect MCP-compatible AI applications to their Appian environment to query application data, analyze business processes, start processes, evaluate expression rules, and call AI agents through natural language.

Benefits of Appian MCP Server

Appian MCP Server provides the following benefits:

  • Standards-based integration: A single MCP server provides a standards-based integration point for any MCP-compatible AI application, eliminating the need for custom API development.
  • Any AI model: Appian MCP Server works with any AI model your connected AI application supports. For a list of verified models, see Supported AI Models for Appian MCP Server.
  • Secure object access: Developers make design objects available as MCP tools through simple configuration, not code. Appian's existing security model applies automatically to all MCP connections without building a separate security layer.
  • Deploy MCP configurations with packages: MCP tool configurations deploy with design objects through packages, eliminating manual re-enablement across environments.

What Appian MCP Server makes available

Appian MCP Server provides external systems with access to the following Appian capabilities:

  • AI agents: Invoke AI agents and check their execution status.
  • Business processes: Query, analyze, and build reports from business processes and datasets in Process HQ.
  • Expression rules: Evaluate expression rules with provided inputs.
  • Process models: Start process models and check their execution status.
  • Record types: Query data from synced record types using SQL.

Developers control which objects are available and to whom. For enablement details, see Make Design Objects Available as MCP Tools.

Appian MCP Server security model

Appian MCP Server supports two authentication methods:

  • User-scoped OAuth 2.0: Each connection authenticates as the signed-in Appian user. Requests run with that user's permissions, and audit logs attribute actions to the individual. Users must be in the Designer role to use this authentication method.
  • Service account API key: All connections authenticate as a single shared service account. The service account's permissions apply to every request.

Regardless of the authentication method, Appian enforces the same security model on all MCP connections, including object security, record-level security, field-level security, and group membership.

User roles in Appian MCP Server

Appian MCP Server involves three roles, each with a distinct workflow.

System administrators

System administrators enable Appian MCP Server and configure authentication in the Admin Console.

After setup, administrators:

Developers

Developers make design objects available as MCP tools through configuration. Developers also write tool descriptions that help AI applications select the right tool, and deploy MCP tool configurations between environments as part of standard design object packages.

End users

End users connect their AI application to Appian MCP Server using configuration details from their user settings. Appian MCP Server works with any supported AI model the connected AI application offers. Once connected, end users interact with Appian data and business logic through natural language.

Appian MCP Server limitations

Appian MCP Server has the following limitations:

  • Synced record types only: Only synced record types can be queried through the data fabric tools. Record types that use direct data access or have a record-level security expression are not accessible.
  • No document upload: External AI applications can retrieve document content but cannot upload documents through MCP.
  • Query result size: Data fabric SQL query results are limited to 128 KB. When a result is larger, the response includes as many complete rows as fit and sets truncated to true. Add filters or select fewer columns to narrow the result.
  • Tool call timeout: Tool calls that run longer than 10 minutes time out.

Appian supports MCP beyond Appian MCP Server. Use the following guidance to determine which option fits your use case:

Option When to use
Appian MCP Server Users already work in a third-party AI application and need to query Appian data, start processes, or call AI agents from that application. The connection flows from the external AI application into Appian.
MCP Tools for AI Agents Appian AI agents need to call tools on external MCP servers, such as Snowflake, GitHub, or Google Drive. The connection flows from Appian out to external systems. Set up with an MCP connected system.
Dev MCP Developers want to use a third-party IDE with AI to create and update Appian design objects. Dev MCP is a community-supported App Market plug-in. For a supported way to build applications using natural language, use Appian Composer.

All three options enforce Appian's existing security model, including object security, record-level security, and group membership.

Feedback