feat: Add Doubleagent - specialized AgentReady development agent

[jeremyeder]

Summary

Implements Doubleagent - a specialized Claude Code agent for AgentReady development tasks.

This agent provides deep knowledge of the AgentReady codebase architecture, assessment workflow, and development patterns to assist with implementing assessors, writing tests, debugging issues, and optimizing performance.

What's New

✅ Comprehensive Agent Specification (.claude/agents/doubleagent.md)

• 607 lines of detailed development guidance
• Deep knowledge of AgentReady architecture and patterns
• Understanding of assessment workflow and scoring algorithm
• Reference implementations and code patterns
• Constitutional principles and best practices
• Performance optimization techniques

Agent Capabilities

The Doubleagent can assist with:

• Implementing new assessors - Following established 5-step pattern
• Expanding stub assessors - 15 stubs ready to implement
• Writing comprehensive tests - With fixtures and >80% coverage target
• Debugging assessment issues - Understanding scoring and finding statuses
• Improving report templates - HTML/Markdown with Jinja2
• Optimizing performance - Caching, lazy loading, early exit strategies

Example Usage

# Implement new assessor
@doubleagent implement new assessor for dependency security scanning

# Enhance existing assessor  
@doubleagent improve TypeAnnotationsAssessor to use AST parsing

# Debug issue
@doubleagent debug why Python type annotation detection is failing

# Expand stub
@doubleagent expand LockFilesAssessor stub into full implementation

Key Knowledge Areas

The agent has comprehensive understanding of:

1. Architecture & Design - Library-first philosophy, strategy pattern, data flow
2. Assessment Workflow - Tier-based weighting (50/30/15/5), scoring algorithm
3. Design Patterns - BaseAssessor, proportional scoring, graceful degradation
4. Reference Implementations - CLAUDEmdAssessor, READMEAssessor, TypeAnnotationsAssessor
5. Test Structure - Unit tests, integration tests, fixtures, coverage targets
6. Design Documents - References to CLAUDE.md, BACKLOG.md, specs/, contracts/

Why "Doubleagent"?

• Double duty: Works for both AI agents AND human developers
• Agent about agents: Helps build a tool that assesses agent-ready codebases
• Meta-level awareness: Understands the principles it's helping implement

Implementation Details

Files Changed:

• .claude/agents/doubleagent.md - New agent specification (607 lines)

Implements: coldstart-prompts/09-agentready-repository-agent.md

Related:

• BACKLOG.md P3 item: "AgentReady Repository Agent"
• Aligns with AgentReady constitution (library-first, TDD when requested)

Test Plan

• ✅ Agent specification created with comprehensive knowledge
• ✅ All markdown linting issues resolved (except MD013 line-length)
• ✅ References all key design documents and schemas
• ✅ Includes concrete code patterns and examples
• ✅ Documents current status and priorities
• 🔲 Manual testing: Invoke @DoubleAgent with development tasks

Checklist

• Code follows library-first architecture
• All linters pass (markdownlint)
• Documentation updated (.claude/agents/)
• Conventional commit format
• No breaking changes

---

🤖 Generated with Claude Code

[github-actions]

🎉 This PR is included in version 1.5.0 🎉

The release is available on GitHub release

Your semantic-release bot 📦🚀