Skip to content

Relevant Code

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
.
├── dto
│   ├── GitHubIssueDTO.java
│   ├── ImportReportDTO.java
│   ├── ImportStatusDTO.java
│   ├── mcpserver
│   │   ├── AskAboutTicketResponseDTO.java
│   │   ├── ChatMessageDTO.java
│   │   ├── ChatRequestDTO.java
│   │   ├── ChatResponseDTO.java
│   │   ├── ChatSessionDTO.java
│   │   ├── SearchResultItemDTO.java
│   │   └── WebSearchResponseDTO.java
├── model
│   ├── ChatMemoryEntity.java
│   ├── ChatSessionEntity.java
│   ├── MessageType.java
├── rest
│   ├── AiResource.java
│   ├── ImportResource.java
├── service
│   ├── ImportService.java
│   ├── mcpserver
│   │   ├── AiService.java
│   │   ├── ChatMemoryProvider.java
│   │   ├── ChatRequestContext.java
│   │   ├── DatabaseChatMemoryStore.java
│   │   └── GoogleSearchApi.java
└── tool
    ├── AskAboutTicketTool.java
    ├── CreateTicketTool.java
    ├── DeleteTicketTool.java
    ├── EditTicketTool.java
    ├── GetAllProjectsTool.java
    ├── GetAllUsersTool.java
    ├── SearchTicketTool.java
    └── WebSearchTool.java

AI Agent Goals

Ticket Management

  • Ticket Discovery – Natural language search across tickets (e.g. "find all open bugs assigned to me blocking the release")
  • Smart Ticket Creation – Draft tickets from a brief prompt, auto-suggesting title, description, priority, size, labels, and acceptance criteria
  • Bulk Operations – Reassign, re-label, close, or update status across multiple tickets in a single command
  • Ticket Summarisation – Summarise long ticket descriptions, comment threads, or entire sprints into a concise overview

Planning & Analysis

  • Implementation Roadmaps – Break a high-level feature or epic into ordered, actionable tickets with dependencies
  • Acceptance Criteria Generation – Derive structured acceptance criteria from a feature description
  • Effort & Risk Assessment – Suggest ticket size/priority and flag potential blockers or missing context

General AI Capabilities

  • Contextual Q&A – Answer questions about the project using existing ticket data (e.g. "what was the reason we dropped X?")
  • Web Search Integration – Look up external context (docs, RFCs, issues) to enrich answers or tickets
  • Natural Language Commands – Perform any supported action via free-text instructions

Non-Functional Requirements

  • Persistent Sessions – Maintain conversation context across multiple interactions so users don't have to repeat themselves
  • Authorisation Enforcement – The agent can only act within the scope of the authenticated user's permissions; no destructive or cross-project actions without explicit confirmation
  • Auditability – Every agent-driven change is attributable and reversible

Atlassian Agentic AI

Atlassian uses a few overlapping names: - Rovo in Jira / Rovo Agents: “agent” experiences embedded into Jira that can help you search, summarize, draft, and (in some setups) take actions like creating/updating work items. (atlassian.com) - Atlassian Intelligence: the umbrella for AI features across Jira/JSM/Confluence (summaries, writing help, knowledge features, etc.). (atlassian.com) - Jira Service Management Virtual Service Agent: a support-focused chatbot for deflecting/automating Tier-1 help requests across channels. (atlassian.com)

Capabilities

1. Speed up reading & triage

  • Summarize issues / long comment threads
  • Summarize linked context

2. Drafting + rewriting work

  • In editors (Jira descriptions, Confluence pages), you can use an agent to draft, rephrase, review, or standardize tone/structure.
  • In JSM specifically, AI can help create/edit knowledge base articles from an issue context.
  • Rovo positions itself as enterprise search + context-aware help across Atlassian content.
  • There are also point features like JQL error fixing to help people who don’t live in JQL daily.

4. Support automation via Virtual Service Agent (ITSM / helpdesk use)

  • A conversational agent for 24/7 self-service that can deflect repetitive Tier-1 questions.

5. Using Jira through ChatGPT

  • Atlassian has been pushing connectors so you can summarize, create/manage issues, and automate workflows in natural language from ChatGPT (via their Rovo MCP ecosystem).