Manage Jira Issues Programmatically

Jira Issue Tool is a LlamaIndex integration that enables operations on Jira issues. It supports searching for issues, creating new ones, adding comments, deleting issues, and modifying key fields like summary, assignee, task status, and due date.

Use this tool when you need to perform Jira issue operations programmatically through LlamaIndex agents, such as creating bugs, reassigning tasks, updating issue statuses, or deleting issues based on search criteria.

Do not use this tool if you need advanced Jira features like sprint management, custom field manipulation beyond the supported operations, or complex JQL query building - it focuses on basic issue CRUD operations only.

You provide Jira server credentials (server URL, email, API token) and a project key to initialize the tool. The agent then accepts commands describing the desired Jira operations.

The tool performs operations such as searching matching issues and updating their status. For example, a command to "find an issue about llamaindex due on Aug 2 2025 and change its status to Done" will search matching issues and update their status.

This tool integrates with:

• Jira: Connects using email and API token authentication to perform issue operations
• LlamaIndex ReActAgent: Works as a tool within LlamaIndex agent frameworks
• OpenAI LLMs: Compatible with OpenAI models (demonstrated with gpt-4o-mini)

from llama_index.tools.jira_issue import JiraIssueToolSpec
from llama_index.llms.openai import OpenAI

llm = OpenAI(model="gpt-4o-mini", api_key=OPENAI_API_KEY, temperature=0.0)

tool_spec = JiraIssueToolSpec(
    server_url=SERVER, email=EMAIL, api_token=API_KEY
)

agent_name = "Jira Issue Agent"
project_key = "YOUR_PROJECT_KEY"
system_prompt = f"""You are a helpful assistant that answers JIRA-related questions with a list of JIRA tools available for use. Your project key is '{project_key}'."""

agent = ReActAgent.from_tools(
    jira_tools,
    llm=llm,
    callback_manager=callback_manager,
    name=agent_name,
    system_prompt=system_prompt,
    verbose=True,
    max_iterations=20,
)

# Multi-step bug issue creation
response = await agent.achat(
    "Create a test bug due by Aug 2 2025 and assign it to John Doe. this bug addresses the wrong temperature setting in the LLMs."
)
print(response)

# Search & Reassign an issue
response = await agent.achat(
    "There's an issue about building a jira agent, assigned to John Doe. Find it and reassign it to Jane Doe."
)
print(response)

This tool is for users who want to perform Jira issue management operations through LlamaIndex agents.