Skip to content
AskVault Docs
Try Free →

Ingest knowledge from Jira

Written by Joseph Russo, Principal Architect, Networks and Tech, AskVault · Reviewed by Allison Chan
Last updated: · 3 min read

What gets imported

For each issue the import matches:

  • Summary and description, including comments (converted from Atlassian's rich-text format to plain text).
  • Issue key and project, included so answers can cite the source issue.

What's not imported:

  • Watchers, version/change history: only the issue's current state.
  • Attachments. File names aren't captured; binary content isn't touched.

Setup walkthrough

About 5 minutes:

  1. Open Knowledge > Add Source > Jira.
  2. Generate an API token at id.atlassian.com if you don't have one.
  3. Enter your Jira URL (e.g. https://yourcompany.atlassian.net), your Atlassian account email, and the API token.
  4. Optionally set a Project Key (e.g. SUPP), or a custom JQL query (overrides the project key if both are set).
  5. Set Max Issues (1–500; defaults to 100).
  6. Click Import Issues.

Use cases

Engineering helpdesk. Engineers ask "Has this bug been reported before?": the bot finds matching imported Jira issues.

Customer support context. Import a status = Done AND labels = customer-facing JQL filter so the bot can reference what's already shipped.

JQL filtering

Use Jira Query Language to scope the import:

  • project = SUPP: only the Support project.
  • status != Closed: only active issues.
  • labels = customer-facing: only issues tagged customer-facing.

JQL is useful to avoid pulling in thousands of internal-only tickets in one import.

Refreshing content

There's no background sync. To pick up issue updates:

  • Run the import again with the same project key or JQL.
  • Issues are matched by URL, so an issue whose content hasn't changed is left alone, and one that changed is re-indexed with the new version.

Limits

  • Issues per import. 500 max, default 100.
  • Attachments, watchers, history. Not imported.
  • Sync. None: re-run the import manually to refresh.

Common pitfalls

"Jira authentication failed." Double-check the email matches the Atlassian account the API token belongs to, and that the token hasn't been revoked.

"Jira search failed: 410." Atlassian retired the Jira search endpoint this import used. An update that moves the import to Atlassian's replacement search is rolling out; until it reaches your account, Jira imports fail with this message. Nothing is wrong with your token or your Jira site.

"Jira search failed: 400: Unbounded JQL queries are not allowed here." Jira's search needs at least one condition, not only an ORDER BY. Add one, such as project = SUPP ORDER BY updated DESC, or leave the JQL field empty to import across all projects.

Stale answers after a Jira update. The import doesn't auto-refresh. Re-run it to pick up the changes.

Too much comment noise. A busy issue's comment thread can overpower retrieval. Use JQL to import a narrower, higher-signal set of issues.

FAQ

Does this use OAuth?

No: email plus an Atlassian API token, entered once per import. This is the same authentication AskVault uses for Confluence, since both are Atlassian products.

Does AskVault keep this in sync automatically?

No, there's no scheduled or webhook-based sync. Re-run the import to refresh; unchanged issues are skipped and changed issues are re-indexed.

Can I index multiple projects?

Yes: run the import once per project key, or write a JQL query that spans multiple projects (e.g. project in (SUPP, ENG)).

Does this affect Jira's own performance?

Reads only, via the standard Jira REST API. No measurable impact under normal use.

Was this page helpful?