Explore every episode of the podcast The Claude Code Changelog

How Claude Code’s experimental /advisor pairs a lower-cost executor with a stronger model that can review the full session and offer guidance. Learn how to enable it, prompt for consultations, inspect advice, and check provider, model-pairing, and cost caveats. We also cover smaller improvements in versions 2.1.301–2.1.303.

Claude Projects is becoming a coordinator for parallel Claude Code cloud sessions: learn how to connect repositories, set project instructions, choose thread models and effort, and manage local-only work. We also cover practical limits and cost traps, including paused sandboxes, pull-request follow-up threads, and usage-limit resumption, plus smaller fixes in 2.1.301 and 2.1.302.

Claude Code 2.1.302 adds CLAUDE_CODE_AUTO_MODE_MODEL, letting users on proxy, Bedrock, or Vertex setups choose the model used by the client-side auto-mode classifier. The episode also covers a breaking default for headless sessions—explicitly set --permission-mode auto in CI—as well as server-side auto-mode compaction and smaller updates to subagents, idle compaction, hooks, image handling, and plugin startup.

Claude Code 2.1.301 adds CLAUDE_CODE_RETRY_WATCHDOG_MAX_USAGE_LIMIT_WAIT_MS, letting CI jobs fail quickly or switch to a fallback model instead of idling through a distant quota reset. Learn how to set it alongside the general retry wait cap, where to scope the watchdog variables, and what “per API request turn” means. Also covered: richer OpenTelemetry tool-decision events, safer worktree cleanup, clearer Grep and Glob errors, and several usability fixes.

Claude Code 2.1.300 adds step-aware workflow routing: mods can rewrite an agent’s prompt and choose its model, letting teams balance cost and capability across a workflow. We cover model fallbacks and allowlists, stricter hook validation, practical safeguards, and several smaller fixes—including clearer cache metrics and improved MCP tool search behavior.

Discover how Claude Code 2.1.299 introduces idle compaction to save you tokens and speed up your workflow after stepping away from the terminal. Lachlan and James break down the 54-minute cache window, configuration settings, and handy updates to hooks and file mentions.

Learn how to control costs and balance performance using the CLAUDE_CODE_WORKFLOW_SUBAGENT_MODEL environment variable introduced in Claude Code 2.1.296. We explore the orchestrator plus swarm pattern, how to configure model overrides without breaking interactive sessions, and other quality-of-life updates in this release.

Learn how to tame 529 overloaded errors in headless Claude Code runs using the newly released watchdog wait caps and retry delays. We break down the ideal CI environment configurations and highlight important quality-of-life fixes in versions 2.1.295 and 2.1.296.

We explore how the new allow_large parameter in Claude Code 2.1.296 handles oversized text files in a single pass while surprisingly cutting token costs. Plus, learn how to trigger it automatically and catch up on key fixes to encodings, PowerShell limits, and hook debugging.

Lachlan and James break down Claude Code 2.1.296, exploring how the new autoCompactWindow setting prevents worker subagents from bloating context windows and running up cache read costs. They also cover practical configuration tips, token limits, and new quality-of-life updates including cheaper Sonnet 5.5 cache reads.

Learn about Claude Code version 2.1.295 and the crucial new onFailure: "block" setting for command and HTTP hooks. We dive into how to prevent broken or missing guard scripts from failing open, plus key caveats to consider before updating your configuration. Show Notes Releases · anthropics/claude-code - GitHub: https://github.com/anthropics/claude-code/releases

Anthropic launched Claude Haiku 5.5 featuring adjustable effort dials, a 1M token context window, and significantly slashed pricing. Learn how to configure custom subagents in Claude Code to take advantage of these massive cost savings while dodging key caveats like token price cliffs and verification quirks.

James and Lachlan explore Claude Code 2.1.292, which introduces a dedicated effort parameter for subagents to dramatically slash token costs in multi-agent workflows. They also cover practical orchestration tips, CI-friendly retry configurations, and critical security patches for UNC paths.

Lachlan and James break down Claude Code versions 2.1.290 and 2.1.291, highlighting new hourly refills for WebSearch budgets and pagination support for truncated WebFetch pages. They also cover critical regression fixes and expanded permission prompts for Bash commands with wildcards.

This episode breaks down the new --max findings flag for Claude Code code review, including how to cap, uncape, and reset findings, plus when to use effort levels and review comments or fixes. It also covers important caveats like sticky settings, background subagent behavior, and related quality-of-life updates in recent Claude Code releases.

This episode covers Claude Code 2.1.289, highlighting fixes to permission rules, sandbox behavior, symlinked file access, and managed machine safeguards. It also breaks down new agent spawn hooks, improved status tracking, and several stability and plugin updates.

This episode covers a fix in Claude Code 2.1.288 that makes path-scoped rules and nested CLAUDE.md files load on Write and Edit, even when a file is created before any read. It also breaks down the new InstructionsLoaded hook details, rule-loading caveats, and a handful of smaller workflow updates in the release.

We break down the new Claude Mods system in Claude Code 2.1.287, including how in-process TypeScript hooks differ from classic JSON hooks, what the built-in You should know mod does, and why security teams are urging caution. Plus: validation tips for third-party mods, team guardrails, 1M context defaults, VS Code updates, and an MCP fix for sign-in prompt changes.

This episode explores Claude Code 2.1.286 and its new verify skill, which runs checks before commits so the agent catches its own mistakes. It also covers related polish like clearer permission prompts, safer transcript actions, background command handling, and a model fallback retry. Show Notes Claude Code changelog - Claude Code Docs: https://code.claude.com/docs/en/changelog

We dig into the latest release updates that make plugin setup far smoother for unattended builds, including install-time config for bundled servers, a command to inspect missing options, and safer ways to pass secrets without exposing them in shell history. Also covered: a fixed bug that silently skipped unconfigured bundled servers, SSH install support that now respects your custom git SSH settings, and a few quality-of-life changes like WebFetch disabling, background command timeouts, and launching Claude Desktop from the terminal.

We unpack Claude Code 2.1.284 and the quiet switch to Sonnet 5.5, from the 1M-token context window and cheaper cache reads to the new 512-token cache minimum. Plus: adaptive thinking defaults, tool-calling changes, keybinding controls, and the token-counting gotchas that could affect your bill.

We dig into Claude Code 2.1.283’s new OTEL logging support, which finally captures MCP, WebFetch, and WebSearch outputs inside trace spans so you can see what your agent actually saw. We also cover the security risks, content limits, and a few welcome MCP quality-of-life fixes like clearer context breakdowns and fewer zombie processes.

Claude Code 2.1.283 changes the default for interactive sessions on Bedrock, Vertex AI, and telemetry-off setups to auto mode, and we break down what that means for approvals, classifier behavior, proxy fallback costs, and how to switch back to manual control. We also cover the new MCP image-saving workflow, prompt file-plus-text options, and a critical PowerShell safety fix for destructive commands.

The hosts break down Claude Code 2.1.283, including the new prompt audit tool that scans CLAUDE.md, skills, subagents, and rules for stale paths, dead commands, and conflicting instructions. They also cover enterprise gateway hint headers and smoother terminal navigation with mouse and page-key support.

We dig into Claude Code 2.1.282, including the new maxProseWidth setting for readable prose on ultrawide monitors, plus self-healing fixes for redacted thinking block errors and broken web search history. We also cover clearer environment diagnostics in claude doctor and /status, along with the new Chrome policy option for managed MCP setups. Show Notes Claude Code changelog - Claude Code Docs: https://code.claude.com/docs/en/changelog

We dig into a frustrating Claude Code bug where MCP tools can silently disappear, then break down how release 2.1.28.1 improves validation by catching dropped .mcp.json entries, undeclared template variables, and insecure HTTP endpoints before they hit production. We also cover a stricter schema gotcha for existing CI pipelines and a handy quality-of-life update for disabling automated Git commit and PR attribution footers.

Claude Code 2.1.280 adds a long-awaited fix for MCP tool descriptions that were silently capped at 2,048 characters, plus a new environment variable to raise or lower the limit as needed. The episode also covers the tradeoffs of larger prompts, a Write tool validation patch, and quality-of-life fixes like improved modal exits and better scrolling.

We break down Claude Code 2.1.277’s enterprise networking fixes, including a new egress-boundary flag that skips local DNS lookups and hands hostname resolution to your forward proxy. We also cover static upstream headers in gateway config, telemetry proxy behavior, streaming gotchas, and a few quality-of-life improvements for worktrees and background tasks.

We break down the latest Claude Code point releases, including the new dynamic boundary for prompt caching that can cut input token costs dramatically, plus practical tips for avoiding cache misses caused by whitespace and line-ending changes. Also covered: automatic prompt sanitization to strip invisible Unicode injection tricks, smoother plugin marketplace registration, and a handy Auto mode status update in slash status.

Claude Code 2.1.277 tackles prompt injection in multi-agent workflows with structural subagent result framing, making it harder for untrusted text to hijack orchestration. The episode also covers better headless error handling, invisible prompt sanitization, and clearer Grep/Glob diagnostics.

Anthropic’s latest Claude Code update moves safety classification checks to the server, cutting token costs for auto mode unless a proxy strips the required headers and forces a fallback. The episode also covers how to diagnose eligibility with /status, disable server checks with an environment variable, and the new AGENTS.md fallback for repo instructions. Show Notes Claude Code 2.1.275 — claude.ai skills and plugins sync to the terminal: https://ai-tldr.dev/releases/anthropic-claude-code-2-1-275/ Auto mode classifier request charges - Claude Code Docs: https://code.claude.com/docs/en/auto-mode-classifier-billing

Anthropic’s latest Claude Code updates make agent steering much smoother, adding a send now shortcut that interrupts a running turn without losing context. The episode also covers account-wide skill and plugin syncing, marketplace install improvements, and a quick hotfix for enterprise gateway and proxy users.

We break down Claude Code 2.1.274’s biggest workflow fixes, including automatic repair for desynced tool calls, cleaner failure handling with slash rewind, and better goal persistence through context compaction. Plus, a look at the new default Model Context Protocol client on Bedrock, Vertex, and Foundry, and how to keep legacy MCP setups running if you need them.

We break down Anthropic’s latest Claude Code update, including opt-in gateway hint headers that let proxies route traffic without inspecting private prompt bodies. Plus, we cover cleaner MCP reconnect notifications, remote session forking, and the return of local default classification for Bedrock and Vertex auto mode.

We break down Anthropic’s Claude Code 2.1.271 update, including the new omitClaudeMd setting that keeps subagents focused by skipping local context files while preserving managed policies. The episode also covers tighter command-level sandboxing, SHA-256 plugin verification, and nicer-looking Markdown artifacts.

We break down the Claude Code 2.1.268 fix for WebFetch hangs, including the new 300-second deadline and the CLAUDE_CODE_WEBFETCH_DEADLINE_MS setting for tuning timeouts in CI. The episode also covers session cleanup for self-hosted runners, JSON output for plugin commands, and tighter local file safety in Cowork sessions.

We break down how Claude Code 2.1.270’s bashEditDiffEnabled surfaces inline diffs for shell edits, helping the agent see file changes instantly without extra git diff calls. The episode also covers a tee-related security fix, a read-only git permission regression, and improved retry handling for more reliable terminal workflows.

Claude Code 2.1.269 brings a dedicated plugin eval CLI for reproducible benchmarking, with scored JSON output and an HTML report to catch regressions before shipping. The episode also covers higher workflow concurrency via CLAUDE_CODE_WORKFLOW_MAX_CONCURRENT_AGENTS and new /output style controls for cleaner remote and headless runs.

We break down a Claude Code bug that caused SessionEnd hooks to be killed after just 1.5 seconds, even when a global timeout was set, leading to truncated teardowns, missing uploads, and corrupted logs. The episode also covers the fix, best practices for per-hook timeouts, and related updates for gateway pricing sync and OpenTelemetry exports.

We break down how Claude Code 2.1.267 helps teams control AI spending with a hard cap on extended thinking across cloud providers, while still letting subagents and custom prompts use lower effort levels when appropriate. The episode also covers the new system prompt snapshot toggle, giving prompt engineers faster live iteration without restarting sessions.

Claude Code 2.1.265 brings hot-reloading for local plugins, letting you point plugin dir at a parent folder and load or unload multiple tools live without restarting your session. The episode also covers symlink safety checks, clearer truncation notices for large tool logs, and a refreshed workflows UI for debugging nested agent activity.

We break down new safeguards for AI coding agents in auto mode, including out-of-bounds read protection, stricter cloud metadata containment, and safer behavior in headless CI runs. The episode also covers terminal usability upgrades and clearer diagnostics for proxy and policy issues.

Claude Code gets a major workflow upgrade with larger inline output buffers that keep long test failures and stack traces in context, cutting out extra file-read tool calls. The episode also covers the token tradeoffs, smarter test filtering, and a handful of CLI polish improvements like subagent prompt files, Bash-style keybindings, and stronger safety checks. Show Notes https://raw.githubusercontent.com/anthropics/claude-code/refs/heads/main/CHANGELOG.md: https://raw.githubusercontent.com/anthropics/claude-code/refs/heads/main/CHANGELOG.md

This episode covers Claude Code 2.1.260’s biggest workflow upgrade: removing the one-hour limit on subagent background tasks, plus new ways to monitor long-running jobs with attach, logs, and stop. It also highlights the new text-based advisor command and safer git worktree deletion prompts that help prevent accidental data loss. Show Notes Claude Code changelog - Claude Code Docs: https://code.claude.com/docs/en/changelog

Claude Code 2.1.261 adds skill doctor to reveal unused loaded skills and their context costs, helping you prune token-heavy workspace setups. The episode also covers larger inline output limits, subagent prompt loading from files, and stronger rm safety checks. Show Notes https://raw.githubusercontent.com/anthropics/claude-code/refs/heads/main/CHANGELOG.md: https://raw.githubusercontent.com/anthropics/claude-code/refs/heads/main/CHANGELOG.md

Anthropic’s latest Claude Code release brings a live split-screen diff panel right into the terminal, making it easier to watch code changes as they happen. The hosts also dig into improved prompt cache miss diagnostics, headless session upgrades, and a security fix for auto-approval in zsh scripts.

We break down Claude Code 2.1.259’s new unattended-host workflow, including --permission-prompts none, how denied prompts keep CI jobs moving, and what to watch for with preapproved tools. The episode also covers the parallel-session state fix, plus quality-of-life updates for GitLab MR commands and plugin validation.

This episode breaks down new Claude Code workflow updates, including how subagent model precedence now works and why the new FORCE environment variable can hard-lock every subagent to a chosen model. It also covers session-level effort controls and new containment rules that add stronger cost and security guardrails for autonomous agents.

This episode breaks down the new direct CLI controls for Claude Code background sessions, including attach, logs, stop, respawn, and rm, plus how they streamline scripting and terminal workflows. It also covers key fixes in versions 2.1.251 and 2.1.252, from git worktree support and macOS task handling to permission saving and cleaner large-error notifications.

This episode breaks down a security bug in Claude Code 2.1.246 where sloppy Bash wildcard rules can accidentally approve dangerous commands, and shows how to rewrite them with explicit subcommands. It also covers the new startup warning, the permissions UI, and handy quality-of-life fixes like completion timestamps and safer terminal diffs.