Documentation - Guardrails Checker

After you install Guardrails Checker, open the admin report from Jira or Confluence settings. You get scoped guardrail tables for your whole site, trend-based breach forecasts from historical snapshots, expandable row deep-dives with remediation guidance, optional manual values the API cannot read, and a one-click CSV export of the current view.

Opening the admin report

Jira admin report — Go to Jira Settings → Apps → Guardrails Checker. You see Jira and Confluence guardrails grouped by administrative scope, with usage percentages and status labels.

Confluence admin report — Go to Confluence Settings → Apps → Guardrails Checker (global page). The same admin report is available when Confluence is on your site.

Jira project report — Open any Jira project and choose Guardrails Checker in the project sidebar for that project’s metrics.

Report layout

At the top, click Refresh instance results to rescan Jira (and Confluence when installed), then sample up to 25 spaces for per-space guardrails. Below that, tables are grouped into four scopes:

Global tenant & site-wide — total spaces, work items, custom fields, and site-wide roles.
Container & per-space — highest value found in any sampled space (fields, workflows, components, and similar limits).
Field, security & permission entities — granular configuration such as select-list options, permission grants, and priorities.
Confluence site-wide — Confluence spaces, pages, export size, and directory signals (when Confluence is on your site).

Each table row shows Guardrail, Usage (% of limit), Current, Limit, and Status. Click any measured row to expand an inline panel with predictive analytics and remediation context.

Predictive analytics & expandable deep-dives

Version 4.2.0 adds Analyze & Forecast for measured guardrails. Click a row (or the ▸ chevron) to open an accordion panel with:

Predictive analytics — Growth velocity (units per month) and estimated days until the guardrail limit is breached, based on linear regression over stored historical snapshots. If growth is stable or shrinking, the panel shows that no breach is projected.
Urgent callout — When a breach is projected within 180 days, the countdown appears in a prominent red callout box.
Top offenders / contextual actions — Sample fields, workflows, or other elements contributing to the warning, each with a Review in Jira Admin link to the relevant administration page.
Remediation playbook — Contextual guidance drawn from the scan (for example, JQL used for work-item counts or sample custom fields on site).
Run Clean Up Wizard (Coming Soon) — A disabled placeholder button for a future guided cleanup flow.

Forecasts improve as history accumulates. Automated hourly snapshots record tenant-level metrics; manual snapshots (see below) add full guardrail rows for richer per-space trends.

Example: expanded row — predictive analytics

	Guardrail	Usage	Current	Limit	Status
▾	Work items (total site count)	82%	14,800,000	18,000,000	Warning

Predictive analytics

Estimated days to breach: 142 days

Growth velocity: +312,500/mo

Top offenders / contextual actions

Site-wide work items — Counted via JQL: created <= now()

→ Review in Jira Admin

Remediation playbook

Work items (approximate count). Site guardrail ~18,000,000.

Status labels

Every row has a status lozenge. Use the legend at the bottom of the admin page to interpret them:

How to read this page

Healthy Below 75% of the limit — no action needed.

Warning Between 75% and 95% — review soon.

Need action At or above 95% — address before performance is impacted.

Awaiting scan Per-space metric — click Refresh to sample spaces and populate Current.

Manual Not available via API — enter the value in the row, then save or export.

Rows in Warning or Need action are highlighted in the table so they stand out from healthy rows.

Scanned, manual, and awaiting scan

Scanned — The app read the value from Jira (or Confluence) APIs. Current shows the number; Usage and Status reflect the limit.
Awaiting scan (gray) — A per-space guardrail that needs a space sample. Click Refresh instance results to populate it.
Manual (light purple) — The API cannot provide this metric reliably. Type a number in the Current column. After you enter a value, Status changes to Healthy, Warning, or Need action based on usage.

Example: mixed scanned and manual rows

Field, security & permission entities

Fine-grained configuration tied to workflows, custom fields, permission schemes, priorities, and security levels.

Guardrail	Usage	Current	Limit	Status
Select-list options per field	80%	8,000	10,000	Warning
Permission grants per scheme	486%	243	50	Need action
Statuses mapped per workflow	—	—	200	Awaiting scan
Work item security levels per space	—	Enter	50	Manual

Manual data entry & CSV export

Some site-wide metrics are not available from the API. Use the Manual data entry section to add:

Active users (this month) — required before you save.
Custom fields override — optional correction when the API count is wrong or missing.
Notes for this export — optional free text (up to 50 characters) included in the CSV.

Your inputs are saved automatically as you type and restored when you reopen the app. After a successful save or export, a grey Last: reference appears next to each field (below the notes field) showing the previous saved value for comparison — that reference is not exported.

Click Save snapshot & download CSV to store a snapshot and download a CSV of the current table view (scanned plus manual values). If your browser blocks the download, use Copy CSV to clipboard when it appears.

Example: manual entry row

Guardrail	Usage	Value	Limit	Status
Notes for this export	—	Q2 migration — billing review Last: Q1 cleanup complete	—	—

Snapshot saved at 6/18/2026, 8:27:02 AM. CSV download started.

CSV contents

The CSV contains one row per guardrail for the current view, including columns for value source (scanned, manual, awaiting, or unset), usage, status, active users, custom fields override, and notes. It does not include historical snapshots or the grey Last: reference labels.

Refreshing the report

Results are cached for faster loads. Use Refresh instance results when you need a fresh pass. The app scans spaces one at a time to stay within Forge limits; progress is shown while sampling runs. Scheduled background refreshes use Forge async events so large instances do not hit single-function timeouts.

Security

Manual number and text fields are validated and sanitized to reduce injection risks. CSV export neutralizes spreadsheet formula prefixes. See our Security Policy for more detail.

What You'll See