Add Coding Agent Governance to Guardrails, Security & Governance

[vanlew1]

I am the author of Coding Agent Governance and am submitting it for the Guardrails, Security & Governance section.

It is an open-source governance layer for bounded AI coding-agent tasks: explicit TaskContract scope checks, task-relevant evidence, stale-verification protection, and structured closure. Its public acceptance evidence is offline and synthetic-only; CLOSED != PRODUCTION_READY.

Repository:
https://github.com/vanlew1/ai-agent-project-governance

Validation:

• The new entry passes the target repository's metadata/schema/rendering checks.
• I ran the full catalog verifier on both the unchanged upstream baseline and this candidate branch.
• The candidate introduces no new verifier failures.
• The remaining failures are pre-existing upstream baseline issues unrelated to this entry:
  1. stale AGENT.md metadata
  2. Claude Agent SDK overview HTTP 404

I did not modify those unrelated entries. Please close or request changes if baseline-red submissions are not acceptable for this catalog.