[{"data":1,"prerenderedAt":45},["ShallowReactive",2],{"$f8LEu8_qCX3Dfmg6AE8EMLhmIgpzmTcNQifDhXVpI7Io":3},{"date":4,"generated_at":5,"picks":6,"candidates_scanned":44,"candidates_scored":28},"2026-04-01","2026-04-01T05:45:00.000000+00:00",[7,20,33],{"rank":8,"title":9,"source":10,"url":11,"category":12,"tldr":13,"score":14,"scores":15,"why":19},1,"v2.1.89","Claude Code Releases (GitHub)","https://github.com/anthropics/claude-code/releases/tag/v2.1.89","Release","- Claude Code v2.1.89 lands with 20+ changes — the headline is a new `\"defer\"` value for PreToolUse hooks: headless sessions can now pause mid-task at any tool call and resume later with `-p --resume`, so long automation pipelines can checkpoint instead of running all-at-once or stopping cold\n- `CLAUDE_CODE_NO_FLICKER=1` is a new env var for flicker-free alt-screen rendering — if you're running Claude Code in tmux or over SSH and seeing visual glitches, set this and they go away\n- Auto mode now shows denied commands in a `/permissions` → Recent tab where you can retry with `r` — previously blocked actions were invisible; now you can see and re-evaluate them\n- Windows users get two critical fixes: Edit/Write were doubling CRLF line endings and stripping Markdown hard line breaks (two trailing spaces), silently corrupting files — update now if you're on Windows\n- `/stats` now correctly counts subagent token usage and preserves data past 30 days — your token history was likely undercounting the whole time",88,{"direct_claude_relevance":16,"practical_utility":17,"novelty":18,"source_credibility":18},33,25,15,"This is the largest single-day Claude Code release in recent memory, with 20+ fixes and features landing simultaneously. Several of them address issues that have been actively frustrating users: the Windows CRLF doubling bug was silently corrupting file edits, the /stats undercounting was hiding real usage costs, and the LSP zombie state crash was forcing full session restarts. The new 'defer' PreToolUse hook is the most architecturally significant addition — it enables genuine pause-and-resume in headless automation without any workarounds.",{"rank":21,"title":22,"source":23,"url":24,"category":12,"tldr":25,"score":26,"scores":27,"why":32},2,"Claude Code GitHub Action v1.0 — Now Generally Available","GitHub anthropics/claude-code-action","https://github.com/anthropics/claude-code-action/releases/tag/v1","- The Claude Code GitHub Action hits v1.0 GA — the old `mode`, `direct_prompt`, and `override_prompt` inputs are replaced with a single `prompt` input, and all CLI options now flow through `claude_args`, making setup dramatically simpler\n- The action now auto-detects interactive vs automation mode (no more manual configuration), and adds native AWS Bedrock and GCP Vertex AI support if you want to route away from the Anthropic API directly\n- Breaking changes from v0.x — review the migration guide before upgrading production workflows",82,{"direct_claude_relevance":28,"practical_utility":29,"novelty":30,"source_credibility":31},30,23,16,13,"Moving from beta to GA is a meaningful signal that Anthropic considers the GitHub Action production-ready — this is the right moment to adopt it or migrate from v0.x before it drifts further from the maintained path. The API surface simplification (everything through `prompt` and `claude_args`) directly resolves the most common source of confusion in the old version, and the Bedrock/Vertex routing options matter for teams with enterprise cloud commitments.",{"rank":34,"title":35,"source":36,"url":37,"category":12,"tldr":38,"score":39,"scores":40,"why":43},3,"claude-agent-sdk-typescript: v0.2.89","GitHub anthropics/claude-agent-sdk-typescript","https://github.com/anthropics/claude-agent-sdk-typescript/releases/tag/v0.2.89","- New `startup()` function pre-warms the CLI subprocess before your first `query()` call — pay the cold-start cost upfront and get ~20x faster first responses when you control the timing\n- New `listSubagents()` and `getSubagentMessages()` functions expose what subagents did inside a session — finally lets you inspect multi-agent execution history programmatically",74,{"direct_claude_relevance":41,"practical_utility":42,"novelty":31,"source_credibility":31},28,20,"The `startup()` addition solves a real latency problem for anyone serving Claude agent responses over an API — cold subprocess boot was a silent 2-5 second penalty on every fresh session, and now you can move it out of the request path entirely. The subagent inspection functions (`listSubagents`, `getSubagentMessages`) fill a genuine observability gap for multi-agent pipelines built on this SDK.",50,1776402243155]