Morning · Tuesday, May 26

AI Brief

171 ready-made Claude Code reasoning skills

  • 12 items
  • 3 axes
  • 3 sources
  • Generated 08:19 UTC

TL;DR

  1. 01
    hn 171 ready-made Claude Code reasoning skills
  2. 02
    hn Isolated Git worktrees for Claude Code
  3. 03
    github Visual guide to Claude Code templates

Stack 1 item · Next.js · Tailwind · GSAP · R3F · shadcn

  1. 01

    Microfrontends routing now applies to vc alias and branch domains

    rss/vercel-blog 8h ago score 7

    Vercel microfrontend routing for aliases — useful for multi-tenant sites.

    vercel.com/changelog

Skills 9 items · Claude Code patterns worth stealing

  1. 01
  2. 02

    Show HN: AWO – Run Claude and Codex in isolated Git worktrees

    hn 12h ago score 8

    Isolated Git worktrees for Claude Code

    github.com/ystepanoff
  3. 03

    luongnv89/claude-howto ★34,218

    github 37m ago score 8

    Visual guide to Claude Code templates

    github.com/luongnv89
  4. 04

    punkpeye/awesome-mcp-servers ★87,900

    github 4h ago score 8

    Curated MCP servers collection

    github.com/punkpeye
  5. 05

    Toward Reliable Design of LLM-Enabled Agentic Workflows: Optimizing Latency-Reliability-Cost Tradeoffs

    rss/arxiv-cs-ai 4h ago score 8

    Optimizing agent workflow latency/cost — actionable patterns.

    arxiv.org/abs
  6. 06

    CODESKILL: Learning Self-Evolving Skills for Coding Agents

    rss/arxiv-cs-ai 4h ago score 8

    Method for self-evolving coding agent skills

    arxiv.org/abs
  7. 07

    Retrying vs Resampling in AI Control

    rss/arxiv-cs-ai 4h ago score 8

    Retry vs resample pattern for Claude Code

    arxiv.org/abs
  8. 08

    Why codex /goal fails on complex workflows: compaction amnesia and context rot

    hn 1h ago score 7

    Diagnoses codex /goal failures, actionable insight

    news.ycombinator.com/item
  9. 09

    Meta-Agent: From Task Descriptions to Verified Multi-Agent Systems

    rss/arxiv-cs-ai 4h ago score 7

    Design pattern for verified multi-agent systems

    arxiv.org/abs

General 2 items · Industry · models · policy

  1. 01

    ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions

    rss/arxiv-cs-cl 4h ago score 8

    Persona drift in long AI coding sessions

    arxiv.org/abs
  2. 02

    Anthropic to release Mythos-class models to the public

    hn 11h ago score 7

    Anthropic public Mythos models

    theregister.com/security