Chart your CMDB on any Jira dashboard — and see the SQL behind every picture. Export Jira Assets to PostgreSQL-ready SQL, CSV and JSON, then query it with real joins, filters and SELECT statements. Runs entirely inside Forge: no external database, no servers, nothing leaves your tenant unless your own BI tool pulls it.
Jira Assets speaks AQL; your BI tools speak SQL. Assets to SQL bridges the two.
It walks your schemas — cross-schema references included — and produces a clean relational model you can download, pull into eazyBI, Power BI or Tableau, or query live on any Jira dashboard. It runs entirely inside your Atlassian tenant on Forge, with read-only access you grant per schema.
PostgreSQL-dialect SQL dump (CREATE TABLE + upserts, delivered zipped), CSV zip with one file per table, and JSON. All stored for re-download — no re-export needed to fetch yesterday's files.
Object types become tables. Reference attributes become {attr}_key / {attr}_name columns plus one level of inlined target attributes, and multi-value references can become proper junction tables, joinable by key.
Background jobs handle up to 50,000 objects per run in a 15-minute processing window, with a live progress bar, per-phase status and a cancel button. Automatic retries on rate limits and transient errors; honest "Partial" labelling when a limit is hit.
Every 6 or 12 hours, daily or weekly — at your wall-clock time in your timezone, daylight-saving safe.
A REST URL serving the latest export as JSON — full, or per-table flat arrays, the shape eazyBI maps directly — or CSV. Bearer-token auth with 256-bit keys, shown once.
Settings lists a ready-made per-table URL for every exported table — paste straight into your BI tool.
See each BI fetch land — time, table, row count — so "did eazyBI actually pull?" is never a mystery.
Read-only CMDB scopes — the app never writes to Assets — and its account reads only the schemas an admin grants, using the standard Assets role model (the same one eazyBI and Automation use).
Data leaves your tenant only when your own BI tool calls the pull endpoint you configured. No developer-operated servers, no analytics.
Optionally restrict who may run exports and change settings. Separately, nothing is queryable on dashboards until an admin exposes it — dashboards reach broad audiences, so gadget exposure is opt-in per table.
A built-in freshness check compares three things — what you can see in Assets, what the app's account is allowed to read, and what your last export actually produced — and names the difference. Missing role grants, object types left out of the configuration, tables that never reached the dashboard store: each is reported with the specific fix, instead of leaving you to guess why a number looks wrong. Every query is labelled with the age of the oldest table it touches, so a partially refreshed store can never masquerade as current.
Comprehensive security and privacy documentation for enterprise reviews. View security page →
The Assets SQL Filter gadget turns your exports into charts on any Jira dashboard — and prints the SQL behind each one. Other Assets apps draw charts; this is the one where anyone can check the query.
Bar, horizontal bar, stacked bar, pie, donut or line — or a single KPI number. Pick the chart first and the builder configures the grouping for you; viewer filters recompute it live.
"Show SQL" prints the exact statement behind any chart — joins included. Reuse it, audit it, or paste it into your own tooling. A picture nobody can check is a picture nobody should trust.
Group by any column and compute a measure — count, distinct count, sum, average, min or max — with top-N ordering by default.
A guided query builder: pick a table, join related tables along real relations in both directions — "what do I point at" and "what points at me" — then choose and rename columns, chain AND/OR conditions, and sort, summarise into groups, and choose how the result is displayed. No SQL required at any point.
Write SELECT statements in a validated dialect: LEFT JOIN, WHERE with AND/OR and parentheses, LIKE/CONTAINS, IS NULL, COUNT/SUM/AVG/MIN/MAX with GROUP BY, ORDER BY on columns or aliases, LIMIT, AS aliases. Statements are parsed and checked against your schema before anything executes — the JQL model, applied to SQL.
Expose up to four columns as viewer-facing dropdowns, fed by live values, or as text search — so viewers slice results without touching the query.
Any gadget result exports to CSV in one click — the full result set, not just the visible page, with the viewer's filters applied. Opens straight into Excel.
Every gadget shows "data as of" its export — gadget data reflects the last export, not live Assets, and says so. Queries spanning tables refreshed at different times report the oldest, labelled "oldest data as of".
Four steps from install to live dashboards.
The pull endpoint gives BI tools a live URL to fetch structured data from. Configure once, stay fresh automatically.
Generate a 256-bit API key in the app settings (shown once), point your BI tool at the endpoint with a Bearer token, and your reports stay fresh automatically. Settings lists a ready-made per-table URL for every exported table.
Authorization: Bearer YOUR_KEY
Serves the latest export as JSON — full, or per-table flat arrays — or CSV. A pull activity log records each fetch with time, table and row count.
Published up front — surprises are worse than limits.
Now live on the Atlassian Marketplace.