Jira update adds Claude Code assignee dropdown

#1 ▶️

NotebookLM 2.0 Update: Analyze Your Entire Business in Minutes! (Full Tutorial)

Helena Liu

#2 𝕏

Andrew Ng says the US Government and Anthropic’s new controls—seen in the Claude Fable 5 release with extra safety guardrails and blocked LLM development—reveal how access to frontier AI can be externally revoked.

#3 𝕏

Harrison Chase recommends ditching the proprietary Claude/Codex harnesses in favor of dcode (Deepagents Code), a model-agnostic harness you can try with FireworksAI’s GLM-5p2 via ``` dcode --model fireworks:accounts/fireworks/models/glm-5p2 ```

#4 𝕏

Santiago unveiled a major Jira update that lets you assign tickets directly to Claude Code via the assignee dropdown or by mentioning it in comments.

#5 📝 Ampcode Chronicle

Custom Agents - You can create custom agents in Amp via plugins and use them as main agents or subagents, include them in tool pipelines or spawn up to 25 worker agents, with each agent getting a custom orb color. The examples show amp.createAgent using model openai/gpt-5.5, registering a focused_review tool and an agent mode, and demonstrate thread APIs—createThread, appendUserMessage (returns immediately), waitForResponse—and an async start_async_review tool that spawns a background thread and returns Started background review in ${thread.id}.

#6 📝 Simon Willison

Sean Lynch - A Hacker News comment from Sean Lynch arguing that the main value of MCP (Model Context Protocol) is isolating authentication outside the agent context, perhaps as a dedicated auth gateway for APIs.

#7 𝕏

clem 🤗 (Clement Delangue) finds cost per task varies ~800× across models—Claude Fable 5 tops performance but costs $31+/task versus ~$0.04 for DeepSeek V4 Flash—while open‐weight GLM-5.2 (max) and DeepSeek V4 Pro (max) deliver the best price/performance (GLM-5.

#8 in

Peter Yang switched from Claude Code to Codex for GPT-5.5’s speed, generous limits, steering controls and best-in-class browser/computer automation. He still uses Claude Code’s Opus frontend and welcomes the ongoing AI competition benefiting builders.

#9 ▶️

My First Winning Agentic AI Trading Strategy On Polymarket

All About AI

#10 𝕏

Boris Cherny applies Claude Code to decipher Linear A, the 3,500-year-old Cretan script, uncovering AI-driven linguistic insights. He’s now hoping the results hold up in peer review.

#11 in

Guillermo Rauch highlights how agents are driving healthy software habits—open APIs, skills documentation, eval-based tests, Unix CLIs, payment/commerce protocols, and widespread markdown/json/html—bringing the original WWW vision to life.

#12 𝕏

Mustafa Suleyman predicts AI-driven healthcare will be the next major product-market-fit explosion, highlighting his collaboration with the Mayo Clinic in a recent discussion with @CoreyNoles.

#13 𝕏

claire vo 🖤 After dozens of interviews, she spots two AI mindsets: companies that are truly cutting-edge but still feel behind, and those using basic AI yet convinced they’ve mastered it.

#14 𝕏

Shreyas Doshi argues you can simplify complex decisions by pinpointing what you’re really selling. Apple sells taste, Amazon convenience, Google utility, Disney nostalgia, Stripe deep care, Anthropic assistance, OpenAI answers, and Starbucks consistency.

#15 𝕏

Santiago has been running the gemma-4:26b model locally on his Mac Studio since April to process private documents, now handling about 60% of his queries.

#16 𝕏

Shreyas Doshi suggests measuring candidates against Claude—ask “are you better than Claude?”—because Claude already outperforms humans at many tasks once considered uniquely human.