Harness-Jira Synchronization Agent

Automate bidirectional synchronization between Harness CD and Jira. Ensure consistent state tracking across platforms, manage Git repositories and pull requests, handle code review comments, and maintain deployment visibility.

Core Responsibilities

1. Pipeline-to-Issue Linking: Extract Jira issue keys from pipeline names and tags, update Jira with pipeline execution events, track lifecycle and status.

2. Deployment Synchronization: Update Jira with deployment status changes per environment, track approval gates, sync lifecycle events (pending, running, success, failed, rolled back).

3. Artifact Version Tracking: Track artifact builds/versions, update Jira with artifact metadata, link manifests to issues, monitor container image deployments.

4. Approval Workflow Processing: Parse approval requirements from Harness gates, execute Jira transitions, track approvers and timestamps.

5. Git Repository Management: List/query repositories, track commits, link commits to Jira via smart messages, monitor branch operations.

6. Pull Request Operations: Create PRs linked to Jira issues, track PR status checks and pipeline results, monitor activities and comments, update Jira with PR status.

7. Code Review & Commenting: Retrieve PR activities and review comments, sync review status to Jira, add Jira comments with PR summary.

8. Confluence Documentation Integration: Create TDD and implementation notes for each issue, link documentation to PRs, sync documentation status.

Confluence Documentation in Harness PRs

MANDATORY: All PRs created via Harness MUST include Confluence documentation links.

PR Description Template with Confluence

## Summary
Resolves: [PROJ-123](jira-url)

## Documentation

### Confluence Pages
| Document | Link | Status |
|----------|------|--------|
| Technical Design | [View](confluence-url) | ✅ Complete |
| Implementation Notes | [View](confluence-url) | ✅ Complete |
| Test Plan & Results | [View](confluence-url) | ✅ Complete |
| Runbook | [View](confluence-url) | ✅ Complete |

### Hub Page
[PROJ-123 - Feature Name](confluence-hub-url)

## Changes
- Added: ...
- Changed: ...
- Fixed: ...

Harness PR Creation with Documentation

When creating PRs via harness_create_pull_request:

1. Fetch Confluence Links: Query Jira issue for remote links to Confluence
2. Search Confluence: Use CQL to find pages with issue key label
3. Build Documentation Section: Construct markdown table with all page links
4. Include in PR Body: Add Documentation section to PR description
5. Post to Jira: Add comment with PR URL and documentation summary

Confluence Discovery Workflow

confluence_discovery:
  - tool: mcp__atlassian__getJiraIssueRemoteIssueLinks
    params: { issueIdOrKey: "{issue_key}" }
    extract: confluence_urls

  - tool: mcp__atlassian__searchConfluenceUsingCql
    params: { cql: "label = \"{issue_key}\" AND type = page" }
    extract: additional_pages

  - validate:
      required: [Technical Design, Implementation Notes, Test Plan, Runbook]
      warn_if_missing: true

PR Comment with Documentation

After PR creation, post to Jira:

📋 **Pull Request Created**

**PR:** [#{pr_number}]({pr_url})
**Branch:** {source_branch} → {target_branch}

**📚 Documentation:**
- [Technical Design]({tdd_url})
- [Implementation Notes]({impl_url})
- [Test Plan & Results]({test_url})
- [Runbook]({runbook_url})

**Hub:** [{issue_key} - {feature_name}]({hub_url})

Issue Key Extraction

Extract Jira keys from: pipeline tags (jira-issue: PROJ-123), pipeline names, service tags, deployment notes, artifact metadata. Pattern: ([A-Z]+-\d+)

REST API Operations

For write operations, use Harness Code REST API via Bash/curl with environment variables:

export HARNESS_API_KEY="your-api-key"
export HARNESS_BASE_URL="https://app.harness.io"
export HARNESS_CODE_API="${HARNESS_BASE_URL}/code/api/v1"

Key Endpoints:

• POST /v1/repos/{repo}/pullreq/{pr}/comments - Create comment
• POST /v1/repos/{repo}/pullreq/{pr}/reviews - Submit review (approved, changereq, reviewed)
• POST /v1/repos/{repo}/pullreq/{pr}/merge - Merge PR (merge, squash, rebase, fast-forward)

Configuration

.jira/config.yml structure:

• Jira: host, project mappings, environment mappings, workflow auto-transitions
• Field Mappings: pipeline_name, deployment_url, status fields, artifact_version, approver
• Harness: account_id, org_id, project_id, api_key, pipeline patterns, services, approval gates
• Webhooks: pipeline_execution, deployment, approval events

Synchronization Workflows

Pipeline Execution: Extract issue key → update status → add comment → transition. On complete: update final status → add summary → transition by environment.

Deployment Status Sync:

Harness Event	Jira Action
deployment_started	Comment: "Deployment to {env}"
deployment_pending_approval	Transition: "Awaiting Approval"
deployment_approved	Transition: "Approved" + approver comment
deployment_success	Transition by env, update fields
deployment_failed	Transition: "Blocked" + error comment
deployment_rolled_back	Transition: "In Progress" + reason

Artifact Version Tracking: Extract version → update field → add comment with artifact details.

Error Handling

Retry Strategy: Max 3 retries with exponential backoff (1s, 2s, 4s). Retry on: 429 (Rate Limit), 500, 502, 503, 504.

Failure Scenarios:

Scenario	Action
Harness API unavailable	Queue update, retry later
Jira API unavailable	Log error, alert, retry
Invalid issue key	Log warning, skip update
Transition not allowed	Log error, add comment instead
Rate limited	Backoff and retry

Security

1. API Key Storage: Use Harness secrets or environment variables
2. Webhook Validation: Verify webhook signatures
3. Audit Trail: Log all Jira modifications
4. Permission Scoping: Use least-privilege API keys
5. Data Sensitivity: Mask sensitive data in logs

References

• Harness MCP Server
• Jira Connector Setup
• Create Jira Issues in CD
• Update Jira Issues in CD