Coding Workflows

Claude Code v2.1.154 introduces Opus 4.8 with dynamic workflows for orchestrating multi-agent tasks, fast mode at reduced costs, and a lean system prompt as default. Key improvements include enhanced safety controls, better plugin management with directory-aware suggestions, and numerous bug fixes for background sessions, terminal rendering, and MCP server handling. The release also deprecates legacy environment variables and improves the user experience across multiple platforms including Bedrock, Vertex, and Chrome integration.

Claude Code v2.1.152 introduces significant enhancements to code review workflows, skill management, and user interface. Key additions include `/code-review --fix` for automatic code improvements, dynamic skill reloading via `SessionStart` hooks, new hook events for message transformation, and improved Vim mode navigation. The release also addresses numerous bug fixes related to terminal rendering, plugin management, MCP server connectivity, and session state handling.

Claude Code v2.1.147 introduces significant improvements to background session management, renames the `/simplify` command to `/code-review` with enhanced bug detection capabilities, and delivers numerous bug fixes across the auto-updater, shell tools, PowerShell integration, and UI rendering. Key enhancements include better session persistence, improved diff performance, and fixes for enterprise login enforcement. The release addresses 40+ issues spanning headless mode compatibility, plugin management, and cross-platform stability.

Claude Code v2.1.146 introduces a command rename from `/simplify` to `/code-review` with effort levels, fixes critical regressions in Windows PowerShell and MCP pagination, and resolves multiple UI/UX issues including full-screen strobing, permission re-prompting, and auto-updater reliability. The release also improves diff rendering performance and fixes several edge cases in background sessions, theme management, and multi-agent configurations.

This guide teaches how to build a grade-and-revise loop using Outcomes in Claude Managed Agents, where a writer agent drafts a cited research brief and a stateless grader independently verifies every URL, quote, and claim against a detailed rubric. The grader provides structured feedback that drives revisions until the brief passes, eliminating manual review cycles. Key techniques include writing specific, actionable rubrics that force concrete evidence, using span.outcome_evaluation_* events to track the loop, and understanding when Outcomes is the right tool for quality assurance.

Google ADK Python v2.0.0-beta.1 marks the transition to Beta phase with significant architectural improvements. Key additions include a full Workflow graph orchestration system with NodeRunner for execution isolation, explicit ReAct loop nodes for agent execution, and Human-in-the-loop resumption capabilities. The release also addresses a critical RCE vulnerability in YAML configuration handling and optimizes performance for single-turn LLM agents.

Claude Code v2.1.111 introduces Claude Opus 4.7 with a new xhigh effort level for tuned speed/intelligence tradeoffs, Auto mode for Max subscribers, and significant UX improvements including interactive effort sliders, comprehensive code review via /ultrareview, and enhanced permission management. The release includes numerous bug fixes for terminal display, LSP diagnostics, plugin handling, and cross-platform compatibility, particularly for Windows PowerShell integration.

Claude Code v2.1.105 introduces significant improvements to worktree management, plugin architecture, and user experience. Key additions include the `path` parameter for EnterWorktree, PreCompact hook blocking, background monitor support for plugins, and the `/proactive` alias for `/loop`. The release focuses on stability with better API stream handling, improved error messaging, and numerous bug fixes across file operations, terminal rendering, and MCP integration.

Claude Code v2.1.101 introduces significant improvements to team collaboration, enterprise security, and user experience. Key additions include a `/team-onboarding` command for generating ramp-up guides, OS CA certificate store trust for enterprise TLS proxies, and auto-creation of cloud environments for remote sessions. The release focuses on enhancing error messaging, fixing critical bugs in session management, and improving plugin/MCP tool reliability across various authentication providers and platforms.

This tutorial demonstrates server-side prompt versioning and rollback for Managed Agents, enabling PMs to update agent prompts without code deployments. It covers creating an agent (v1), evaluating it against a labeled test set, shipping an updated prompt (v2), detecting performance regressions, and rolling back by pinning sessions to a specific version. The workflow replaces traditional code-based prompt management with immutable versioned prompts that can be quickly reverted if issues arise.

This tutorial demonstrates building a webhook-triggered SRE incident response agent using Claude Managed Agents that automatically investigates production alerts, consults runbooks, proposes infrastructure fixes via pull requests, and gates merging behind human approval. The agent combines built-in sandbox tools (bash, read, edit) with custom tools for PR management and human-in-the-loop approval, providing complete audit trails in the Anthropic Console. The example uses mocked PagerDuty, GitHub, and Datadog integrations to focus on agent patterns, with guidance for swapping in real services.

This tutorial introduces the Claude Managed Agents API by walking through a practical debugging workflow where an agent iteratively fixes failing tests in a Python package. It covers the three core resources (Agent, Environment, Session), file mounting, and the streaming event loop pattern. The example demonstrates how agents autonomously discover the iterate-observe-fix loop by running tests, reading failures, editing code, and repeating until all assertions pass.

The Claude Agent SDK provides built-in session management functions to build conversation history sidebars without writing custom parsers. This cookbook demonstrates how to list, read, rename, tag, and fork sessions stored as JSONL transcripts on disk. By leveraging SDK primitives like list_sessions(), get_session_messages(), and rename_session(), developers can implement session browsers for desktop apps, IDE extensions, and chatbots without managing file I/O directly.

Claude Code v2.1.91 introduces MCP tool result persistence overrides supporting up to 500K characters, adds shell execution disabling options, enables multi-line prompts in deep links, and allows plugins to ship executables. The release includes critical fixes for transcript chain breaks, terminal key handling, plan mode persistence in remote sessions, and JSON schema validation, plus performance improvements and enhanced API guidance.

Claude Code v2.1.83 introduces significant improvements to policy management, environment reactivity, security, and user experience. Key additions include a drop-in directory for managed settings, new hook events for environment changes, transcript search functionality, and enhanced subprocess credential scrubbing. The release addresses numerous stability issues, performance regressions, and UI glitches while improving startup latency and memory usage across various platforms.

Claude Code v2.1.77 introduces significant improvements to token limits for Claude Opus and Sonnet models (up to 128k), enhances sandbox filesystem controls with `allowRead` settings, and fixes numerous critical bugs affecting bash commands, memory management, API tracking, and terminal UI interactions. The release also improves performance with faster startup times and session resumption, better plugin validation, and refined agent communication workflows.

This cookbook demonstrates how to connect Claude agents to external systems using MCP (Model Context Protocol) servers for GitHub monitoring and CI workflows. The guide covers integrating Git and GitHub MCP servers to enable agents to interact with repositories, manage workflows, and perform observability tasks. By leveraging MCP servers, agents gain access to specialized tools for Git operations, GitHub platform integration, and CI/CD monitoring without relying on command-line interfaces.

This cookbook teaches how to build a research agent using Claude Agent SDK with WebSearch tool for autonomous information gathering and synthesis. The guide demonstrates creating a functional research agent in just a few lines of code, then progresses to production improvements including conversation memory, system prompts for specialized behavior, and multimodal research capabilities. The agent autonomously decides when and how to search, follows promising leads, and synthesizes findings without predefined workflows.

This cookbook guide teaches developers how to create, deploy, and manage custom skills for Claude that extend its capabilities with organization-specific workflows and domain knowledge. Custom skills are specialized expertise packages bundled as markdown files, scripts, and resources that codify organizational knowledge, ensure consistency, and automate complex workflows while maintaining privacy. The guide covers skill architecture, SKILL.md requirements, progressive disclosure for token optimization, and provides utility functions for skill management including creation, listing, and deletion.

This cookbook demonstrates automatic context compaction for managing token limits in long-running agentic workflows. It shows how the Claude Agent Python SDK can automatically compress conversation history when token usage exceeds a threshold, enabling tasks to continue beyond the 200k token context limit. The example uses a customer service agent processing support tickets, where each ticket requires multiple tool calls that accumulate in conversation history. By implementing context compaction with the compaction_control parameter, agents can maintain focus and efficiency across many iterations without manual context management.