For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Connect
BlogHelp CenterAPI PricingSign up
OverviewElevenCreativeElevenAgentsElevenAPIReception AIAPI referenceChangelog
OverviewElevenCreativeElevenAgentsElevenAPIReception AIAPI referenceChangelog
  • Get started
    • Overview
    • Quickstart
  • Configure
    • Overview
    • Voice & language
    • Knowledge base
    • Tools
    • Personalization
    • Authentication
      • Cal.com
      • Calendly
      • Cursor
      • Exa
      • HubSpot
      • Jira
      • Parallel
      • Salesforce
      • ServiceNow
      • Slack
      • Twilio
      • Zendesk
  • Deploy
    • Overview
    • Environment variables
    • WhatsApp
    • Batch calls
  • Monitor
    • Overview
    • Users
    • Testing
    • Experiments
    • Versioning
    • Conversation Analysis
    • Analytics
    • Real-time monitoring
    • OpenTelemetry traces
    • Privacy
    • Cost optimization
    • CLI
  • Advanced
    • Events
    • Custom models
    • LLM cascading
    • Post-call webhooks
  • Resources
    • UI components
  • Guides
    • Chat Mode
    • Burst pricing
    • ElevenLabs' docs agent
    • Scaling user interviews
    • Simulate Conversations
LogoLogo
Login
Login
Connect
BlogHelp CenterAPI PricingSign up
On this page
  • Overview
  • Setup
  • Identifying callers
  • How it works
  • Example system prompt
  • Available tools
  • Useful links
ConfigureIntegrations

Jira

Connect your ElevenLabs agents with Jira Cloud
Was this page helpful?
Previous

Parallel

Connect your ElevenLabs agents with Parallel web search
Next
Built with

Overview

Connect your ElevenLabs AI agents with Jira Cloud to manage issues, track projects, and automate support workflows. This integration enables your agents to create and update issues, transition issues through workflows, link related tickets for escalation, and search across projects.

Setup

This integration uses a Jira API token with Basic authentication. You need your Atlassian account email, an API token, and your Jira Cloud subdomain.

1

Generate an API token

Go to Atlassian API token management and click Create API token. Copy the token - you won’t be able to see it again.

2

Find your Jira subdomain

Your subdomain is the first part of your Jira Cloud URL. For example, if your Jira is at mycompany.atlassian.net, your subdomain is mycompany.

3

Connect in ElevenLabs

In the ElevenLabs integration setup, enter:

  • Email - your Atlassian account email
  • API Token - the token you generated in step 1
  • Jira Subdomain - your subdomain from step 2

API tokens expire after one year by default. If your agent stops being able to access Jira, generate a new token and update the connection.

Identifying callers

The jira_search_issues tool requires a labels parameter to scope searches to a specific user’s tickets. This assumes your Jira workflow applies a label to each issue that identifies the caller — for example, converting an email like jane.smith@example.com into a label like jane-smith-at-example-com.

To automate this, configure the labels field as a dynamic variable that receives the caller’s identifier from your application. This way the agent does not need to ask the caller for their email on every call.

Jira labels cannot contain spaces. A common convention is to replace @ with -at- and . with - (e.g., jane-smith-at-example-com). The jira_search_issues tool matches the label exactly, so the value provided must match the label on the issue.

How it works

Support ticket creation
Service desk resolution
1

Caller identification

The agent identifies the caller by email address or another identifier configured in the labels field.

2

Gather details

The agent collects information about the issue from the caller - what happened, severity, and any relevant context.

3

Create the issue

The agent creates a Jira issue in the appropriate project with the right issue type, priority, and description.

4

Confirm with the caller

The agent reads back the created ticket number so the caller has a reference for follow-up.

Example system prompt

View example system prompt
# Personality
You are a helpful IT support agent responsible for managing support tickets in Jira. Be friendly, professional, and efficient.
# Environment
You operate in a support setting via voice or chat, where you help users report issues, check ticket status, and manage existing requests.
# Tone
Begin by asking how you can help. If the caller has an existing issue, ask whether they have the ticket number. If they are reporting a new issue, collect details about the problem.
Ask one question at a time and wait for the response before proceeding.
# Goal
For new issues:
- Collect a clear description of the problem
- Determine severity and urgency
- Create the ticket in the appropriate project with the right issue type and priority
- Read back the ticket number
For existing issues:
- If the caller provides a ticket number, look it up by key
- If the caller wants to check on their issues, search using the labels field
- Provide status updates
- Add comments or update priority when requested
- Always confirm before making changes
# Guardrails
- Do not discuss topics outside of IT support.
- Never expose raw Jira API responses, internal IDs, or technical field names to the caller.
- Always confirm with the caller before creating, updating, or transitioning issues.

Available tools

ToolDescription
jira_create_issueCreate an issue with project, summary, issue type, and optional fields like priority.
jira_get_issueRetrieve details for a specific issue by key.
jira_search_issuesSearch for issues by label, project, status, priority, type, assignee, text, or date range.
jira_update_issueUpdate fields on an existing issue - priority, assignee, labels, or description.
jira_add_commentAdd a plain-text comment to an issue.
jira_get_transitionsList available workflow transitions for an issue.
jira_transition_issueMove an issue to a new workflow status using a transition ID from jira_get_transitions.
jira_link_issuesLink two issues together (e.g., “Blocks”, “Duplicate”, “Relates”).
jira_list_projectsList available projects and their keys.
jira_get_issue_typesList available issue types (e.g., Bug, Task, Story).
jira_get_prioritiesList available priority levels.

Issue descriptions and comments are plain text only. The integration converts them into Jira’s Atlassian Document Format (ADF) automatically. Markdown and HTML are not supported.

Useful links

  • Jira Cloud REST API documentation
  • Manage API tokens