Jira Integration
Connect Jira as a context source for the AI Documentation Agent to search issues with JQL, read issue details, and keep your docs aligned with what your engineering team is building.
Overview
Use Jira as a live context source for the AI Documentation Agent. The Writer and Answer agents read Jira issues at runtime through an AI subagent, so they work from current issue data instead of a copied snapshot.
The agent queries Jira on demand with JQL, or Jira Query Language, to find the exact issues it needs for a task. This works well for changelogs, release notes, feature documentation, and migration guides that need to reflect what your engineering team has shipped or is actively building.
The agent reads Jira live at runtime through your OAuth connection. It does not import or sync Jira issues into your Documentation.AI workspace.
Before you start
Make sure you have the access and project setup required to enable Jira for an AI agent.
- Organization or workspace admin role in Documentation.AI
- Atlassian account with Jira access to the projects and issues you want the agent to read
- At least one documentation project where you want to enable the Jira connector
Connect Jira
Connect Jira once at the organization level, then enable the connector in each documentation project that should use it.
Open organization integrations
Go to Settings → Organization Settings → Integrations.
Find Jira, then click Connect.
Complete the Atlassian OAuth flow
Sign in to your Atlassian account and approve access.
After you finish the OAuth flow, Documentation.AI redirects you back to the dashboard.
Success check: the integrations page shows Jira as connected.
Enable Jira for a documentation project
Go to Settings → Integrations inside the specific documentation project that should use Jira.
Turn on the Jira agent connector for that project.
Success check: the project now shows Jira as an enabled context source for the AI agent.
Verify the active connection
Confirm the connection status shows as active.
You should also see the connected Jira site name and URL, which confirms the agent can use that Jira instance at runtime.
The Jira OAuth connection is organization-level, but the Jira agent connector toggle is enabled per project. Connect once, then turn it on only in the projects that need Jira context.
How the AI agent uses Jira
The Jira connector gives the AI agent a focused set of tools for searching issues, reading details, and understanding release activity.
| Tool | What it does |
|---|---|
listProjects | Lists all Jira projects accessible to the connected account. |
searchIssues | Searches issues using JQL by project, status, fix version, label, or full-text. |
getIssue | Fetches full details of a single issue by key such as DOCS-123, including status, assignee, priority, description, fix versions, components, linked issues, and parent epic. |
getIssueComments | Reads comments on an issue for team discussions, decisions, and workarounds. |
listBoards | Lists Jira boards, optionally scoped to a project. |
listSprints | Lists sprints on a board, filterable by state such as active, closed, or future. |
listReleases | Lists a project's fix versions or releases with names, release dates, and status. |
The agent decides which tools to call based on your prompt. If you ask for release notes, it may search issues by fix version and then read issue details or comments before drafting documentation.
Example prompts
Use prompts like these when you want the AI agent to pull context from Jira while generating or updating docs.
- Search Jira for issues with fixVersion 2.1.0 in the DOCS project and write release notes
- Find all issues in the API project closed in the last sprint and create a changelog entry
- Read the details of DOCS-456 and create a feature documentation page based on the issue description and comments
- List all releases in the MOBILE project and create a release history page with dates and status
- Search for issues labeled breaking-change and update the migration guide accordingly
Troubleshooting
Fix most Jira connection problems by checking access, project settings, and the scope of the OAuth connection.
Linear support is coming soon as an additional issue tracker context source with a similar setup flow.