For the complete documentation index, see llms.txt. This page is also available as Markdown.

Decision Register Overview

Purpose

Explain how Decision Register creates an audit-style history from Jira issue change events.

Who it is for

  • Jira administrators and project leads

  • PMO, governance, and audit teams

  • Engineering managers reviewing delivery drift

Prerequisites

  • Decision Register app installed on Jira Cloud

  • Access to Jira projects and issue history data

Product page

Website: Decision Register for Jira

Steps

  1. Open Decision Register from the Jira Global Page or Admin Page.

  2. Configure run filters for scope and period.

  3. Run a scan to generate decision entries from issue history.

  4. Review grouped or table output, counts, and metadata.

  5. Export results for reporting or audit sharing.

Capability summary

  • Forge app exposed as Jira Global Page and Admin Page

  • Generates audit-style register from issue change history

  • Detects estimate, priority, scope or sprint, rollback or reopen, assignee, due date, and release or fix version changes

  • Entry fields include issue key, summary, when, who, from, and to

  • Supports run filters: project selection, time range, max issues scanned, issue types, and only-my-changes

  • Estimate field handling supports auto-detect and optional customfield override

  • Results in grouped or table views with per-group visibility controls

  • Type counts and run metadata included

  • Issue links in results for fast verification

  • Export options: CSV plus fallback copy-to-clipboard

  • Per-user preferences persisted in Forge storage with reset-to-defaults

  • Resilience includes retries for rate-limit and server errors, first-run init message, and partial-result warnings

Rovo positioning

Decision Register complements Atlassian AI/Rovo by adding structured, auditable change evidence for governance and operational review.

Common issues

  • Empty output is often caused by narrow filter scope or short time windows.

  • Large scans can return partial warnings; use run metadata and rerun with adjusted limits.

Last updated