In case anyone else searches on this, I've had pretty good luck simply asking Claude to rely on zen as much as possible, and then to incorporate the zen capabilities into the constitution. So far, the spec-kit commands are respecting the instructions, and of course greatly enhancing the code quality.

https://github.com/BeehiveInnovations/zen-mcp-server

Collaboration & Planning (Enabled by default)

chat - Brainstorm ideas, get second opinions, validate approaches
thinkdeep - Extended reasoning, edge case analysis, alternative perspectives
planner - Break down complex projects into structured, actionable plans
consensus - Get expert opinions from multiple AI models with stance steering
Code Analysis & Quality

debug - Systematic investigation and root cause analysis
precommit - Validate changes before committing, prevent regressions
codereview - Professional reviews with severity levels and actionable feedback
analyze (disabled by default - enable) - Understand architecture, patterns, dependencies across entire codebases
Development Tools (Disabled by default - enable)

refactor - Intelligent code refactoring with decomposition focus
testgen - Comprehensive test generation with edge cases
secaudit - Security audits with OWASP Top 10 analysis
docgen - Generate documentation with complexity analysis
Utilities

challenge - Prevent "You're absolutely right!" responses with critical analysis
tracer (disabled by default - enable) - Static analysis prompts for call-flow mapping

Claude's response: The constitution now mandates collaborative AI development to ensure higher code quality, better edge case
coverage, and multiple perspectives on all development decisions.