Atlassian Integration Skill

Overview

This skill provides comprehensive integration with the Atlassian ecosystem including Jira, Confluence, and Bitbucket. Use it for project management automation, documentation workflows, and source control operations.

Authentication Setup

API Token Authentication

# Set environment variables
export ATLASSIAN_DOMAIN="your-domain.atlassian.net"
export ATLASSIAN_EMAIL="your-email@example.com"
export ATLASSIAN_API_TOKEN="your-api-token"

# Test authentication
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/myself" | jq '.displayName'

Generate API Token

1. Go to https://id.atlassian.com/manage-profile/security/api-tokens
2. Click "Create API token"
3. Label it descriptively (e.g., "agent-studio-automation")
4. Copy and store securely in .env

Jira Operations

Issue Management

# Get issue details
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/issue/PROJECT-123" | jq '.'

# Create issue
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/issue" \
  -d '{
    "fields": {
      "project": {"key": "PROJECT"},
      "summary": "Issue summary",
      "description": {
        "type": "doc",
        "version": 1,
        "content": [{"type": "paragraph", "content": [{"type": "text", "text": "Description"}]}]
      },
      "issuetype": {"name": "Task"},
      "priority": {"name": "Medium"},
      "assignee": {"accountId": "account-id"}
    }
  }'

# Update issue
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X PUT \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/issue/PROJECT-123" \
  -d '{"fields": {"summary": "Updated summary", "priority": {"name": "High"}}}'

# Search issues with JQL
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/search?jql=project=PROJECT+AND+status='In+Progress'&maxResults=50" | jq '.issues[].key'

# Add comment
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/issue/PROJECT-123/comment" \
  -d '{
    "body": {
      "type": "doc",
      "version": 1,
      "content": [{"type": "paragraph", "content": [{"type": "text", "text": "Comment text"}]}]
    }
  }'

Issue Transitions

# Get available transitions
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/issue/PROJECT-123/transitions" | jq '.transitions[] | {id, name}'

# Transition issue (e.g., move to Done)
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/issue/PROJECT-123/transitions" \
  -d '{"transition": {"id": "31"}}'

Sprint Management

# Get board sprints (requires Jira Software)
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/rest/agile/1.0/board/BOARD_ID/sprint?state=active" | jq '.'

# Move issues to sprint
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/rest/agile/1.0/sprint/SPRINT_ID/issue" \
  -d '{"issues": ["PROJECT-123", "PROJECT-124"]}'

Epic and Story Hierarchy

# Get epics for project
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/search?jql=project=PROJECT+AND+issuetype=Epic" | jq '.issues[] | {key: .key, summary: .fields.summary}'

# Link issues (parent/child, blocks, etc.)
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/rest/api/3/issueLink" \
  -d '{
    "type": {"name": "Blocks"},
    "inwardIssue": {"key": "PROJECT-123"},
    "outwardIssue": {"key": "PROJECT-124"}
  }'

Confluence Operations

Page Management

# Get space pages
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/wiki/rest/api/content?spaceKey=SPACE&type=page&limit=25" | jq '.results[] | {id, title}'

# Create page
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/wiki/rest/api/content" \
  -d '{
    "type": "page",
    "title": "Page Title",
    "space": {"key": "SPACE"},
    "body": {
      "storage": {
        "value": "<p>Page content in HTML</p>",
        "representation": "storage"
      }
    }
  }'

# Update page (requires current version number)
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X PUT \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/wiki/rest/api/content/PAGE_ID" \
  -d '{
    "type": "page",
    "title": "Updated Title",
    "version": {"number": 2},
    "body": {
      "storage": {
        "value": "<p>Updated content</p>",
        "representation": "storage"
      }
    }
  }'

# Search pages
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/wiki/rest/api/content/search?cql=space=SPACE+AND+title~'search+term'" | jq '.'

Confluence Macros and Templates

# Create page from template
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/wiki/rest/api/content" \
  -d '{
    "type": "page",
    "title": "Meeting Notes 2024-01-15",
    "space": {"key": "SPACE"},
    "ancestors": [{"id": "PARENT_PAGE_ID"}],
    "body": {
      "storage": {
        "value": "<ac:structured-macro ac:name=\"info\"><ac:rich-text-body><p>Meeting notes content</p></ac:rich-text-body></ac:structured-macro>",
        "representation": "storage"
      }
    }
  }'

Bitbucket Operations

Repository Management

# List repositories
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://api.bitbucket.org/2.0/repositories/WORKSPACE" | jq '.values[] | {name: .name, clone: .links.clone}'

# Create pull request
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST \
  -H "Content-Type: application/json" \
  "https://api.bitbucket.org/2.0/repositories/WORKSPACE/REPO/pullrequests" \
  -d '{
    "title": "Feature: Add new capability",
    "description": "PR description",
    "source": {"branch": {"name": "feature-branch"}},
    "destination": {"branch": {"name": "main"}},
    "reviewers": [{"account_id": "reviewer-account-id"}]
  }'

# Get PR status
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://api.bitbucket.org/2.0/repositories/WORKSPACE/REPO/pullrequests/PR_ID" | jq '{state, title, author}'

MCP Server Configuration

Setup atlassian MCP Server

{
  "mcpServers": {
    "atlassian": {
      "command": "npx",
      "args": ["-y", "@atlassian/mcp-atlassian"],
      "env": {
        "ATLASSIAN_DOMAIN": "your-domain.atlassian.net",
        "ATLASSIAN_EMAIL": "your-email@example.com",
        "ATLASSIAN_API_TOKEN": "your-api-token"
      }
    }
  }
}

Add to .claude/settings.json under mcpServers for agent-studio integration.

Available MCP Tools (after setup)

• jira_get_issue — Retrieve issue details by key
• jira_create_issue — Create new Jira issue
• jira_update_issue — Update issue fields
• jira_search_issues — JQL-based search
• jira_transition_issue — Move issue through workflow
• confluence_get_page — Retrieve Confluence page
• confluence_create_page — Create new page
• confluence_update_page — Update existing page
• confluence_search — CQL-based content search

Cross-Product Workflows

Dev-to-Jira Automation

# Extract Jira issue key from git branch and update status
BRANCH=$(git branch --show-current)
ISSUE_KEY=$(echo "$BRANCH" | grep -oP '[A-Z]+-[0-9]+')

if [ -n "$ISSUE_KEY" ]; then
  # Move to In Progress when branch is created
  TRANSITION_ID=$(curl -s -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
    "https://$ATLASSIAN_DOMAIN/rest/api/3/issue/$ISSUE_KEY/transitions" | \
    jq -r '.transitions[] | select(.name == "In Progress") | .id')

  curl -s -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
    -X POST -H "Content-Type: application/json" \
    "https://$ATLASSIAN_DOMAIN/rest/api/3/issue/$ISSUE_KEY/transitions" \
    -d "{\"transition\": {\"id\": \"$TRANSITION_ID\"}}"

  echo "Moved $ISSUE_KEY to In Progress"
fi

Confluence Doc Generation from Jira Sprint

# Generate sprint report page in Confluence
SPRINT_ID="123"
SPACE_KEY="TEAM"

# Get sprint issues
ISSUES=$(curl -s -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  "https://$ATLASSIAN_DOMAIN/rest/agile/1.0/sprint/$SPRINT_ID/issue" | \
  jq -r '.issues[] | "- [\(.key)] \(.fields.summary) — \(.fields.status.name)"' | \
  sed 's/&/\&amp;/g; s/</\&lt;/g; s/>/\&gt;/g')

# Create Confluence page with sprint summary
curl -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
  -X POST -H "Content-Type: application/json" \
  "https://$ATLASSIAN_DOMAIN/wiki/rest/api/content" \
  -d "{
    \"type\": \"page\",
    \"title\": \"Sprint $SPRINT_ID Report\",
    \"space\": {\"key\": \"$SPACE_KEY\"},
    \"body\": {
      \"storage\": {
        \"value\": \"<h2>Sprint Issues</h2><p>$ISSUES</p>\",
        \"representation\": \"storage\"
      }
    }
  }"

Error Handling

# Robust API call with error handling
atlassian_api() {
  local method="$1"
  local endpoint="$2"
  local data="$3"

  local response
  response=$(curl -s -w "\n%{http_code}" \
    -u "$ATLASSIAN_EMAIL:$ATLASSIAN_API_TOKEN" \
    -X "$method" \
    -H "Content-Type: application/json" \
    ${data:+-d "$data"} \
    "https://$ATLASSIAN_DOMAIN$endpoint")

  local http_code
  http_code=$(echo "$response" | tail -1)
  local body
  body=$(echo "$response" | head -n -1)

  if [[ "$http_code" -ge 400 ]]; then
    echo "ERROR: HTTP $http_code — $(echo "$body" | jq -r '.errorMessages[]? // .message // "Unknown error"')" >&2
    return 1
  fi

  echo "$body"
}

# Usage
atlassian_api GET "/rest/api/3/issue/PROJECT-123"
atlassian_api POST "/rest/api/3/issue" '{"fields": {...}}'

Rate Limiting

• Jira Cloud: 10 requests/second per user
• Confluence: 10 requests/second per user
• Implement exponential backoff for 429 responses

# Retry with backoff
retry_api() {
  local max_attempts=3
  local attempt=1
  local wait=1

  while [ $attempt -le $max_attempts ]; do
    if atlassian_api "$@"; then
      return 0
    fi
    echo "Attempt $attempt failed. Retrying in ${wait}s..." >&2
    sleep $wait
    wait=$((wait * 2))
    attempt=$((attempt + 1))
  done

  echo "All attempts failed" >&2
  return 1
}

Best Practices

1. Use JQL efficiently — Filter server-side, not client-side. Add project/assignee/sprint constraints.
2. Batch operations — Use bulk update endpoints when modifying multiple issues.
3. Store issue keys — Always reference issues by key (PROJECT-123), not internal ID.
4. ADF for descriptions — Jira v3 API uses Atlassian Document Format (ADF), not plain text or wiki markup.
5. Webhook subscriptions — For real-time updates, register webhooks instead of polling.
6. Service accounts — Use dedicated service account API tokens for automation, not personal tokens.
7. Field customization — Custom fields use IDs like customfield_10001 — query /rest/api/3/field to discover them.

Related Skills

• devops — CI/CD pipeline integration
• github-ops — GitHub alternative for source control
• azure-devops — Microsoft DevOps platform alternative