Connect AI to Your Jira Projects

Transform how you manage and track your work by connecting Claude, Cursor AI, and other AI assistants directly to your Jira projects, issues, and workflows. Get instant project insights, streamline issue management, and enhance your team collaboration.

• Ask AI about your projects: "What are the active issues in the DEV project?"
• Get issue insights: "Show me details about PROJ-123 including comments"
• Track project progress: "List all high priority issues assigned to me"
• Manage issue comments: "Add a comment to PROJ-456 about the test results"
• Search across projects: "Find all bugs in progress across my projects"
• Create and update issues: "Create a new bug in the MOBILE project"

• Developers who need quick access to issue details and development context
• Project Managers tracking progress, priorities, and team assignments
• Scrum Masters managing sprints and workflow states
• Team Leads monitoring project health and issue resolution
• QA Engineers tracking bugs and testing status
• Anyone who wants to interact with Jira using natural language

Get up and running in 2 minutes:

Generate a Jira API Token:

1. Go to Atlassian API Tokens
2. Click Create API token
3. Give it a name like "AI Assistant"
4. Copy the generated token immediately (you won't see it again!)

# Set your credentials
export ATLASSIAN_SITE_NAME="your-company"  # for your-company.atlassian.net
export ATLASSIAN_USER_EMAIL="your.email@company.com"
export ATLASSIAN_API_TOKEN="your_api_token"

# List your Jira projects
npx -y @aashari/mcp-server-atlassian-jira get --path "/rest/api/3/project/search"

# Get details about a specific project
npx -y @aashari/mcp-server-atlassian-jira get --path "/rest/api/3/project/DEV"

# Get an issue with JMESPath filtering
npx -y @aashari/mcp-server-atlassian-jira get --path "/rest/api/3/issue/PROJ-123" --jq "{key: key, summary: fields.summary, status: fields.status.name}"

Add this to your Claude configuration file (~/.claude/claude_desktop_config.json):

{
  "mcpServers": {
    "jira": {
      "command": "npx",
      "args": ["-y", "@aashari/mcp-server-atlassian-jira"],
      "env": {
        "ATLASSIAN_SITE_NAME": "your-company",
        "ATLASSIAN_USER_EMAIL": "your.email@company.com",
        "ATLASSIAN_API_TOKEN": "your_api_token"
      }
    }
  }
}

Restart Claude Desktop, and you'll see the jira server in the status bar.

Most AI assistants support MCP. Install the server globally:

npm install -g @aashari/mcp-server-atlassian-jira

Then configure your AI assistant to use the MCP server with STDIO transport.

Create ~/.mcp/configs.json for system-wide configuration:

{
  "jira": {
    "environments": {
      "ATLASSIAN_SITE_NAME": "your-company",
      "ATLASSIAN_USER_EMAIL": "your.email@company.com",
      "ATLASSIAN_API_TOKEN": "your_api_token"
    }
  }
}

Alternative config keys: The system also accepts "atlassian-jira", "@aashari/mcp-server-atlassian-jira", or "mcp-server-atlassian-jira" instead of "jira".

This MCP server provides 5 generic tools that can access any Jira API endpoint:

Projects:

• /rest/api/3/project/search - List all projects (paginated, recommended)
• /rest/api/3/project - List all projects (non-paginated, legacy)
• /rest/api/3/project/{projectKeyOrId} - Get project details

Issues:

• /rest/api/3/search/jql - Search issues with JQL (use jql query param). IMPORTANT: /rest/api/3/search is deprecated!
• /rest/api/3/issue/{issueIdOrKey} - Get issue details
• /rest/api/3/issue - Create issue (POST)
• /rest/api/3/issue/{issueIdOrKey}/transitions - Get/perform transitions

Comments:

• /rest/api/3/issue/{issueIdOrKey}/comment - List/add comments
• /rest/api/3/issue/{issueIdOrKey}/comment/{commentId} - Get/update/delete comment

Worklogs:

• /rest/api/3/issue/{issueIdOrKey}/worklog - List/add worklogs
• /rest/api/3/issue/{issueIdOrKey}/worklog/{worklogId} - Get/update/delete worklog

Users & Statuses:

• /rest/api/3/myself - Get current user
• /rest/api/3/user/search - Search users (use query param)
• /rest/api/3/status - List all statuses
• /rest/api/3/issuetype - List issue types
• /rest/api/3/priority - List priorities

By default, all responses use TOON (Token-Oriented Object Notation) format, which reduces token usage by 30-60% compared to JSON. TOON uses tabular arrays and minimal syntax, making it ideal for AI consumption.

To use JSON instead: Add --output-format json to CLI commands or set outputFormat: "json" in MCP tool calls.

Example TOON vs JSON:

TOON: key|summary|status
      PROJ-1|First issue|Open
      PROJ-2|Second issue|Done

JSON: [{"key":"PROJ-1","summary":"First issue","status":"Open"},
       {"key":"PROJ-2","summary":"Second issue","status":"Done"}]

All tools support optional JMESPath (jq) filtering to extract specific data:

# Get just project names and keys
npx -y @aashari/mcp-server-atlassian-jira get \
  --path "/rest/api/3/project/search" \
  --jq "values[].{key: key, name: name}"

# Get issue key and summary
npx -y @aashari/mcp-server-atlassian-jira get \
  --path "/rest/api/3/issue/PROJ-123" \
  --jq "{key: key, summary: fields.summary, status: fields.status.name}"

For large API responses (>40k characters ≈ 10k tokens), responses are automatically truncated with guidance. The complete raw response is saved to /tmp/mcp/mcp-server-atlassian-jira/<timestamp>-<random>.txt for reference.

When truncated, you'll see:

• A truncation notice with the raw file path
• Suggestions to refine your query with better filters
• Percentage of data shown vs total size

Ask your AI assistant:

• "List all projects I have access to"
• "Show me details about the DEV project"
• "What projects contain the word 'Platform'?"

Ask your AI assistant:

• "Find all high priority issues in the DEV project"
• "Show me issues assigned to me that are in progress"
• "Search for bugs reported in the last week"
• "List all open issues for the mobile team"

Ask your AI assistant:

• "Get full details about issue PROJ-456 including comments"
• "What's the current status and assignee of PROJ-123?"
• "Display all comments on the authentication bug"

Ask your AI assistant:

• "Add a comment to PROJ-456: 'Code review completed, ready for testing'"
• "Comment on the login issue that it's been deployed to staging"

The CLI mirrors the MCP tools for direct terminal access:

# GET request (returns TOON format by default)
npx -y @aashari/mcp-server-atlassian-jira get --path "/rest/api/3/project/search"

# GET with query parameters and JSON output
npx -y @aashari/mcp-server-atlassian-jira get \
  --path "/rest/api/3/search/jql" \
  --query-params '{"jql": "project=DEV AND status=\"In Progress\"", "maxResults": "10"}' \
  --output-format json

# GET with JMESPath filtering to extract specific fields
npx -y @aashari/mcp-server-atlassian-jira get \
  --path "/rest/api/3/issue/PROJ-123" \
  --jq "{key: key, summary: fields.summary, status: fields.status.name}"

# POST request (create an issue)
npx -y @aashari/mcp-server-atlassian-jira post \
  --path "/rest/api/3/issue" \
  --body '{"fields": {"project": {"key": "DEV"}, "summary": "New issue title", "issuetype": {"name": "Task"}}}'

# POST request (add a comment)
npx -y @aashari/mcp-server-atlassian-jira post \
  --path "/rest/api/3/issue/PROJ-123/comment" \
  --body '{"body": {"type": "doc", "version": 1, "content": [{"type": "paragraph", "content": [{"type": "text", "text": "My comment"}]}]}}'

# PUT request (update issue - full replacement)
npx -y @aashari/mcp-server-atlassian-jira put \
  --path "/rest/api/3/issue/PROJ-123" \
  --body '{"fields": {"summary": "Updated title"}}'

# PATCH request (partial update)
npx -y @aashari/mcp-server-atlassian-jira patch \
  --path "/rest/api/3/issue/PROJ-123" \
  --body '{"fields": {"summary": "Updated title"}}'

# DELETE request
npx -y @aashari/mcp-server-atlassian-jira delete \
  --path "/rest/api/3/issue/PROJ-123/comment/12345"

Note: All CLI commands support:

• --output-format - Choose between toon (default, token-efficient) or json
• --jq - Filter response with JMESPath expressions
• --query-params - Pass query parameters as JSON string

1. Check your API Token permissions:

   • Go to Atlassian API Tokens
   • Make sure your token is still active and hasn't expired

2. Verify your site name format:

   • If your Jira URL is https://mycompany.atlassian.net
   • Your site name should be just mycompany

3. Test your credentials:

   npx -y @aashari/mcp-server-atlassian-jira get --path "/rest/api/3/myself"

1. Check the API path:

   • Paths are case-sensitive
   • Use project keys (e.g., DEV) not project names
   • Issue keys include the project prefix (e.g., DEV-123)

2. Verify access permissions:

   • Make sure you have access to the project in your browser
   • Some projects may be restricted to certain users

1. Try different search terms:

   • Use project keys instead of project names
   • Try broader search criteria

2. Check JQL syntax:

   • Validate your JQL in Jira's advanced search first

1. Restart Claude Desktop after updating the config file
2. Verify config file location:
   • macOS: ~/.claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json

If you're still having issues:

1. Run a simple test command to verify everything works
2. Check the GitHub Issues for similar problems
3. Create a new issue with your error message and setup details

Your Atlassian account needs:

• Access to Jira with the appropriate permissions for the projects you want to query
• API token with appropriate permissions (automatically granted when you create one)

Currently, this tool only supports Jira Cloud. Jira Server/Data Center support may be added in future versions.

Your site name is the first part of your Jira URL:

• URL: https://mycompany.atlassian.net -> Site name: mycompany
• URL: https://acme-corp.atlassian.net -> Site name: acme-corp

Any AI assistant that supports the Model Context Protocol (MCP):

• Claude Desktop
• Cursor AI
• Continue.dev
• Many others

Yes! This tool:

• Runs entirely on your local machine
• Uses your own Jira credentials
• Never sends your data to third parties
• Only accesses what you give it permission to access

Yes! Use JQL queries for cross-project searches. For example:

npx -y @aashari/mcp-server-atlassian-jira get \
  --path "/rest/api/3/search/jql" \
  --query-params '{"jql": "assignee=currentUser() AND status=\"In Progress\""}'

Version 3.2.1 (December 2025):

• Added TOON output format for 30-60% token reduction
• Implemented automatic response truncation for large payloads (>40k chars)
• Raw API responses saved to /tmp/mcp/mcp-server-atlassian-jira/ for reference
• Updated to MCP SDK v1.23.0 with modern registerTool API
• Fixed deprecated /rest/api/3/search endpoint (now use /rest/api/3/search/jql)
• Updated all dependencies to latest versions (Zod v4.1.13, Commander v14.0.2)

• Node.js: 18.0.0 or higher
• MCP SDK: v1.23.0 (uses modern registration APIs)
• Jira: Cloud only (Server/Data Center not supported)

This server follows the 5-layer MCP architecture:

1. CLI Layer - Human interface using Commander.js
2. Tools Layer - AI interface with Zod validation
3. Controllers Layer - Business logic and orchestration
4. Services Layer - Direct Jira REST API calls
5. Utils Layer - Cross-cutting concerns (logging, formatting, transport)

Enable debug logging by setting the DEBUG environment variable:

# In Claude Desktop config
{
  "mcpServers": {
    "jira": {
      "command": "npx",
      "args": ["-y", "@aashari/mcp-server-atlassian-jira"],
      "env": {
        "DEBUG": "true",
        "ATLASSIAN_SITE_NAME": "your-company",
        "ATLASSIAN_USER_EMAIL": "your.email@company.com",
        "ATLASSIAN_API_TOKEN": "your_api_token"
      }
    }
  }
}

Debug logs are written to ~/.mcp/data/mcp-server-atlassian-jira.<session-id>.log

Check raw API responses: When responses are truncated, the full raw response is saved to /tmp/mcp/mcp-server-atlassian-jira/<timestamp>-<random>.txt with request/response details.

Version 3.0 replaces 8+ specific tools with 5 generic HTTP method tools. If you're upgrading from v2.x:

Before (v2.x):

jira_ls_projects, jira_get_project, jira_ls_issues, jira_get_issue,
jira_create_issue, jira_ls_comments, jira_add_comment, jira_ls_statuses, ...

After (v3.0+):

jira_get, jira_post, jira_put, jira_patch, jira_delete

Migration examples:

• jira_ls_projects -> jira_get with path /rest/api/3/project/search
• jira_get_project -> jira_get with path /rest/api/3/project/{key}
• jira_get_issue -> jira_get with path /rest/api/3/issue/{key}
• jira_create_issue -> jira_post with path /rest/api/3/issue
• jira_add_comment -> jira_post with path /rest/api/3/issue/{key}/comment
• jira_ls_statuses -> jira_get with path /rest/api/3/status

Benefits of v3.0+:

• Full access to any Jira REST API v3 endpoint (not just predefined tools)
• JMESPath filtering for efficient data extraction
• Consistent interface across all HTTP methods
• TOON format for 30-60% token savings
• Automatic response truncation with raw file logging

Need help? Here's how to get assistance:

1. Check the troubleshooting section above - most common issues are covered there
2. Visit our GitHub repository for documentation and examples: github.com/aashari/mcp-server-atlassian-jira
3. Report issues at GitHub Issues
4. Start a discussion for feature requests or general questions

Made with care for teams who want to bring AI into their project management workflow.