Unified Ticket Fetcher

Unified Ticket Fetcher Skill

Stateless atomic operation to securely query multiple project management and enterprise ticketing platforms (Jira, Plane, ServiceNow), aggregate assigned work items, resolve credentials gracefully, and produce a single normalized task schema.

Prerequisites

• unified-task-tracker — for orchestrating multi-provider authentication and issuing high-fidelity API requests to ticketing backends.

Steps

Step 1: poll_assigned_tickets

Identify and query active ticketing endpoints for the current user session:

• Locate active credentials and connection properties for:
  • Jira: Resolve URL, username, and API token.
  • Plane: Resolve workspace name, project IDs, and API key.
  • ServiceNow: Resolve instance URL, client ID, and credential pairs.
• Execute parallel or sequential queries to each active ticketing platform:
  • Call Jira endpoint (e.g., query assignee = currentUser() AND statusCategory != Done).
  • Call Plane endpoint (e.g., retrieve active issues assigned to user profile identifier).
  • Call ServiceNow incident tracker (e.g., retrieve incidents where assigned_to matches user ID and state is not resolved/closed).
• Output parameters:
  • raw_payloads: Dictionary mapping provider name (jira, plane, servicenow) to lists of raw ticket metadata.

Step 2: aggregate_ticket_payloads [depends_on: poll_assigned_tickets]

Collate raw records, manage connection failures, and extract core ticket attributes:

• Handle execution faults:
  • If a specific provider API request fails (e.g., due to timeout or bad credentials), capture the exception, register a warning diagnostic, and continue processing other available backends (graceful degradation).
• Consolidate raw ticket listings:
  • Merge the raw JSON arrays into a single in-memory collection.
  • Apply global filter logic to remove completed, archived, or resolved issues.
• Output parameters:
  • aggregated_list: Consolidated list of active raw ticket metadata objects.
  • failed_providers: List of providers that failed to respond.

Step 3: normalize_tasks [depends_on: aggregate_ticket_payloads]

Transform heterogeneous ticket schemas into a single, standardized, unified task list:

• Standardize fields into the target schema:
  • id: Normalized ticket identifier string (e.g. JIRA-101, PLANE-402, SN-INC00123).
  • source: Provider identifier ("jira", "plane", or "servicenow").
  • title: Short title or summary string.
  • description: Main description body, stripped of platform-specific HTML/formatting where appropriate.
  • status: Normalized workflow state ("TODO", "IN_PROGRESS", "BLOCKED").
  • priority: Standardized severity level ("LOW", "MEDIUM", "HIGH", "CRITICAL").
  • due_date: ISO-8601 due date string (null if unspecified).
  • created_date: ISO-8601 creation timestamp.
• Sort and deduplicate:
  • Remove duplicate records if any exist.
  • Sort the task collection by priority descending, then by created_date ascending.
• Output parameters:
  • status: "SUCCESS" or "FAILED"
  • normalized_tasks: List of standard task objects containing all normalized keys.
  • task_count: Total number of active tickets retrieved.