Lawful AI Daily Brief — 2026-04-29

🛠️ Tool Updates

• Claude Code kept polishing operator life: OAuth retry loop fix in v2.1.123, better MCP duplicate handling, and alwaysLoad for MCP servers.
• Codex CLI is on espresso shots again: 0.126.0-alpha.9 → .13 with heavier MCP/subagent import workflows and tighter provider/tool gating.
• Practical takeaway: orchestration is getting stronger and stricter — great for compliance-minded automation.

💡 Tip of the Day

If your legal workflow keeps “forgetting” tools, force-load your legal MCP server and stop context roulette 🎯

{
  "mcpServers": {
    "legaldb": {
      "url": "http://localhost:8787",
      "alwaysLoad": true
    }
  }
}

⚖️ Legal x AI Watch

• Fresh regulation-focused repo activity:
  • artvana/global-ai-atlas — machine-readable global AI legislation atlas.
  • ThatLadyKatieGovernance/Ai-governance-analyses — governance framework breakdowns and strategy notes.
  • Alvoradozerouno/GENESIS-v10.1 — sovereign-AI compliance framing (EU AI Act + financial regs).
• Fast compliance win: log model + tool + retrieved sources per legal answer so audits don’t become archaeological digs later.

📚 Fresh Papers

• Navigating Global AI Regulation: A Multi-Jurisdictional Retrieval-Augmented Generation System
  https://arxiv.org/abs/2604.25448v1
• CORAL: Adaptive Retrieval Loop for Culturally-Aligned Multilingual RAG
  https://arxiv.org/abs/2604.25676v1
• Carbon-Taxed Transformers: A Green Compression Pipeline for Overgrown Language Models
  https://arxiv.org/abs/2604.25903v1

📈 Trending Repos

• Significant-Gravitas/AutoGPT ⭐ 183k — still the big autonomous-agent gravity well.
• langgenius/dify ⭐ 139k — production agent workflow platform momentum continues.
• langchain-ai/langgraph ⭐ 30k — graph-style resilient agent architecture keeps climbing.

🎤 Standup One-Liner

“Yesterday we tightened our legal AI stack so it’s faster in practice and much easier to defend in a compliance review.”

---

Repo: https://github.com/laugustyniak/lawful-ai-staging