Add experimental multi-agent autonomous Claude code development workflow

[tjementum]

Summary & Motivation

Add an experimental agentic workflow system that enables autonomous feature implementation through coordinated Claude Code agents. The system introduces specialized agents (tech-lead, coordinator, engineers, reviewers) that work together to implement features from PRD creation through code review and commit, maintaining enterprise-grade code quality.

• Add Claude Code agent definitions and system prompts for 9 agent types (tech-lead, coordinator, 3 engineers, 3 reviewers, pair-programmer, parallel-tool-runner)
• Add workflow commands for task implementation, reviews, and feature coordination with structured STEP 0 preparation
• Add mode commands for activating tech-lead, coordinator, and loading workflow documentation
• Add MCP tools for agent delegation (start_worker_agent), task completion (complete_work), problem reporting (report_problem), and model switching
• Add Developer CLI commands for managing worker-host Claude Code sessions (claude-agent, init-task-manager)
• Add lifecycle utilities for agent coordination, message file protocol, and session management
• Add logging infrastructure for debugging agent workflows
• Add git helper methods for detecting recent activity to support idle monitoring
• Add Claude Code hooks to enforce MCP tool usage and prevent dangerous git operations
• Improve commit workflow with automatic sync-ai-rules execution for .claude changes
• Update AI rules with clarifications (TelemetryEventsCollectorSpy usage, repository DTO restrictions, varchar sizing)
• Fix modal dialog pattern to use dynamic isDismissable based on mutation state
• Fix end-to-end test terminology from "step decorators" to "step wrappers"
• Fix form validation rules to remove deprecated FormErrorMessage component
• Update README with multi-agent workflow documentation and expanded .claude directory structure

The workflow supports Linear integration by default, with flexibility to use Azure DevOps, Jira (via MCP), or file-based Markdown tracking in .workspace/task-manager/.

Checklist

• I have added tests, or done manual regression tests
• I have updated the documentation, if necessary

[linear]

PP-641 Multi-agent workflow

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud