Workflow Automation

Corsa Workflows let compliance teams automate repeatable operational steps without code. Build a workflow visually, choose what should trigger it, add blocks for decisions and actions, then activate it when ready. Use workflows to reduce manual handoffs across onboarding, transaction monitoring, alert triage, case operations, and periodic reviews.

How It Works

Choose a trigger: an event in Corsa, such as a new alert, updated case, created transaction, or client update.
Add blocks to decide what happens next: update records, create follow-up items, branch by condition, or notify the right people.
Save changes as a draft until the workflow is ready.
Activate the workflow so it starts running automatically.
Review each execution in the workflow history, including step-by-step outcomes and linked results.

Workflow Builder

The builder gives your team a visual canvas for designing automation logic. Every workflow starts with a trigger and then flows through one or more blocks. Available workflow blocks include:

Record - Create or update Corsa records, including alerts, cases, clients, and transactions.
Action - Trigger a customer interaction, such as an email-based request for information.
Copilot - Run an AI analysis using Corsa Copilot. Write a prompt, optionally enable tools, and use the output in downstream steps.
Notify - Send an in-app, email, or Slack notification.
Branch - Split the workflow into conditional paths based on alert, case, or client data.

You can rename workflows, edit blocks, add parallel branches, and discard unsaved changes before publishing.

Keep workflows small and focused. A workflow that owns one operational outcome is easier to review, test, and audit.

Triggers

Workflows can start from Corsa events or from a recurring schedule.

Event Triggers

Event-based workflows run when matching activity happens in Corsa. Supported event families include:

Individual client created or updated
Corporate client created or updated
Transaction created or updated
Alert created or updated
Case created or updated

Each trigger can include filters, so the workflow only runs when the event matches your criteria. For example, you can trigger a workflow only for high-priority transaction monitoring alerts, clients with a specific risk level, or cases in a specific stage.

Scheduled Triggers

Scheduled workflows run on a recurring interval. Use them for periodic review, stale-case checks, recurring client controls, or other time-based operations. Schedules support minute, hour, day, and week intervals, with a minimum interval of 5 minutes. Scheduled workflows can target individual or corporate clients and can include filters to narrow the population.

Draft, Active, Inactive, and Archived

Every workflow has a lifecycle:

Draft - Editable, but not running yet.
Active - Live and eligible to run automatically.
Inactive - Paused. You can reactivate it later.
Archived - Retained for history, but no longer editable.

Drafts are useful while building or updating a workflow. When you activate a workflow, Corsa validates the workflow structure first, so incomplete or invalid workflows do not go live.

Execution History

Each workflow keeps a history of its runs. Use the execution list to see status, start and completion time, duration, creator, failed step, and generated results. Execution statuses include Pending, Running, Completed, Failed, Cancelled, and Timed out. Open an execution to inspect the timeline. The timeline shows each step, whether it succeeded, failed, skipped, or is still running, and any output or error details available for that step. When a workflow creates or updates a linked result, such as an alert, Corsa shows that result directly in the execution detail so analysts can jump from the run history to the affected record.

Copilot Block

The Copilot block brings AI into your workflows. Write a prompt describing the analysis you need, and Corsa runs it automatically each time the workflow executes. The AI output is available to downstream blocks — use it to drive branch decisions, populate record fields, or enrich notifications.

Configuring a Copilot Block

Every Copilot block needs a prompt — the instruction the AI will follow at runtime. Prompts support smart data variables ({{entity.fieldName}}, {{trigger.entityType}}, etc.) so the AI receives context from the triggering event automatically.

Prompt

The instruction the AI follows. Use smart data variables to inject runtime context. For example:

Analyze the transaction history for {{entity.overview_name}} and determine
whether recent activity patterns suggest structuring behavior.

Model

Choose which AI model runs the analysis. Options include Claude (Anthropic) and GPT (OpenAI) models. The default is Claude Sonnet 4.6.

Output schema

Define a JSON Schema for structured output. When set, the AI returns data matching your schema instead of free-form text. This is useful when downstream blocks need to read specific fields from the result.

{
  "type": "object",
  "properties": {
    "isSuspicious": { "type": "boolean" },
    "reasoning": { "type": "string" },
    "riskScore": { "type": "number" }
  },
  "required": ["isSuspicious", "reasoning"]
}

Tools

Enable tool groups so the AI can query live data during execution. Available tool groups:

Entity query - Look up client, alert, case, and transaction data.
Metrics - Query compliance metrics and statistics.
Screening - Run PEP and sanctions screening lookups.
Web search - Search the web for public information.

Tools are read-only — the AI cannot modify any data through tools.

Context from previous steps

Select upstream workflow blocks whose outputs should be included as additional context for the AI. For example, if a risk assessment block runs before the Copilot block, you can feed the risk score and reasoning into the AI prompt automatically.

Temperature

Controls how deterministic or creative the AI response is. Lower values (closer to 0) produce more consistent, factual output. Higher values (closer to 1) allow more variation. For compliance workflows, a low temperature (0.0–0.2) is recommended.

System instructions

Additional instructions that shape how the AI behaves. These are appended to Corsa’s built-in compliance workflow instructions, which ensure outputs are clean and machine-readable.

Max tokens

Limits the length of the AI response (1–32,000 tokens). Set this to control cost and ensure outputs stay concise.

Using Copilot Output in Downstream Blocks

The AI result is available to all blocks that come after the Copilot block in the workflow:

Branch blocks can route based on fields in the AI output (e.g., branch on isSuspicious being true or false).
Record blocks can use AI output values in field updates via smart data variables.
Notify blocks can include AI-generated text in notification messages.
Other Copilot blocks can reference the output as context for multi-step AI reasoning.

Use an output schema when you need to branch on AI results or map specific values to record fields. Free-form text works well for notifications and descriptions.

Copilot in Scheduled Workflows

When a scheduled workflow matches multiple entities, the Copilot block runs independently for each entity with its own context. Each execution is isolated — the AI sees only the data for the entity it is analyzing.

Common Use Cases

Alert triage automation

When a high-priority alert is created, assign ownership, update priority, notify the compliance team, and branch based on alert category.

Case operations

When a case changes stage, update related fields, notify reviewers, or trigger customer follow-up.

Client review controls

Run scheduled checks over individual or corporate clients and start review actions when risk or activity criteria match.

Transaction follow-up

When a transaction is created or updated, branch by amount, status, type, or risk context and create downstream compliance work.

AI-powered alert enrichment

When a new alert is created, use a Copilot block to analyze the entity’s history, generate a risk summary, and auto-populate the alert description before routing to a reviewer.

Best Practices

Start with a narrow trigger and add filters early.
Use branches for meaningful business decisions, not cosmetic structure.
Keep record updates explicit so analysts can understand what changed.
Review execution history after activation to confirm the workflow behaves as expected.
Pause inactive workflows instead of deleting operational history.

Case Management

Notifications

AI Copilot

Data Store

Workflow Automation

How It Works

Workflow Builder

Triggers

Event Triggers

Scheduled Triggers

Draft, Active, Inactive, and Archived

Execution History

Copilot Block

Configuring a Copilot Block

Using Copilot Output in Downstream Blocks

Copilot in Scheduled Workflows

Common Use Cases

Alert triage automation

Case operations

Client review controls

Transaction follow-up

AI-powered alert enrichment

Best Practices

​How It Works

​Workflow Builder

​Triggers

​Event Triggers

​Scheduled Triggers

​Draft, Active, Inactive, and Archived

​Execution History

​Copilot Block

​Configuring a Copilot Block

​Using Copilot Output in Downstream Blocks

​Copilot in Scheduled Workflows

​Common Use Cases

Alert triage automation

Case operations

Client review controls

Transaction follow-up

AI-powered alert enrichment

​Best Practices

How It Works

Workflow Builder

Triggers

Event Triggers

Scheduled Triggers

Draft, Active, Inactive, and Archived

Execution History

Copilot Block

Configuring a Copilot Block

Using Copilot Output in Downstream Blocks

Copilot in Scheduled Workflows

Common Use Cases

Best Practices