Claude Agent SDK
PulseAugur coverage of Claude Agent SDK — every cluster mentioning Claude Agent SDK across labs, papers, and developer communities, ranked by signal.
- 2026-08-31 product_launch Anthropic released version 0.2.149 of the Claude Agent SDK for Python. source
- 2026-07-08 product_launch Anthropic's Claude Agent SDK has introduced a new plugin system to simplify the integration of custom tools and components. source
- 2026-06-14 product_launch Anthropic is implementing a new credit system for the Claude Agent SDK starting June 15. source
- 2026-05-23 product_launch Anthropic is implementing a separate billing structure for the Claude Agent SDK and programmatic CLI usage. source
- 2026-05-14 product_launch Anthropic introduces a new credit system for the Claude Agent SDK and resumes third-party tool access.
7 day(s) with sentiment data
-
Build Personal AI Assistant with Claude Code and Agent SDK
A guide details how to build a personal AI assistant using Anthropic's Claude Code and Claude Agent SDK, leveraging the Model Context Protocol (MCP) for integration with services like Gmail, Google Calendar, and Slack. …
-
Agent harnesses, frameworks, and MCP: Understanding AI agent architecture layers
The article distinguishes between three distinct layers in AI agent architecture: agent harnesses, agent frameworks, and the Model Context Protocol (MCP). Agent harnesses, like those in OpenAI's Codex and Anthropic's Cl…
-
AI models struggle with JSON structure, prompting schema-based fixes
A content pipeline encountered issues with AI models generating JSON output that, while syntactically valid, did not conform to the expected data structure. This led to failures in downstream validation processes, parti…
-
Anthropic simplifies agent design with single-model commerce reference
Anthropic has released a reference implementation for commerce agents, showcasing a simplified architecture that relies on a single Claude model in a standard agent loop. This approach eschews the complex orchestration …
-
AI sub-agents often gain broader tool permissions than their parents
This analysis investigates how sub-agents inherit permissions from their parent agents across five different AI frameworks: LangGraph, Deep Agents, CrewAI, Claude Code, and the Claude Agent SDK. The findings reveal that…
-
Anthropic's Claude Fable 5.1 adds "preserved thinking" to block distillation
Anthropic has introduced a new security measure in its Claude Fable 5.1 model to prevent "distillation," a technique where models are trained on the outputs of other models. This new feature, called "preserved thinking,…
-
LangGraph, OpenAI, and Claude Agent SDKs Compared for AI Agent Development
This article compares three leading agent frameworks: LangGraph, OpenAI Agents SDK, and Claude Agent SDK. It analyzes their distinct approaches to managing agent loops and solving similar problems with different core pr…
-
Anthropic releases Claude Agent SDK v0.2.149 for Python
Anthropic has released version 0.2.149 of its Claude Agent SDK for Python. This update is available on the Python Package Index, allowing developers to install it using pip.
-
Ollama integrates with ChatGPT Desktop; Claude Code sees multiple updates
Ollama has released version 0.34.0-rc4, enabling users to integrate Ollama models directly into ChatGPT Desktop. This update improves structured output performance on Apple Silicon and adds support for OpenAI-compatible…
-
AWS Bedrock AgentCore Evaluations uses OpenTelemetry for framework-agnostic AI agent assessment
Amazon Bedrock AgentCore Evaluations offers a new way to assess AI agent frameworks by decoupling evaluation from specific framework choices. It leverages OpenTelemetry, a vendor-neutral instrumentation standard, to col…
-
Thomson Reuters develops proprietary AI model; New York tech talent surpasses San Francisco
Thomson Reuters has developed its own AI model tailored for legal, tax, and compliance tasks, while still integrating Anthropic's Claude Agent SDK into its CoCounsel Legal product. Separately, New York's tech workforce …
-
Casebook MCP makes AI agent incident knowledge queryable
A new system called Casebook MCP has been developed to make AI agent incident knowledge queryable. It functions as a remote Model Context Protocol (MCP) server, allowing agents like Claude Code or Cursor to access a reg…
-
AI Harnesses, Not Models, Are the New Tech Battleground
The AI race has shifted from developing larger models to perfecting the "harness" that surrounds them, according to a dev.to post. This harness, encompassing prompts, tools, and control flow, has become the new battlegr…
-
Anthropic's Claude Cookbook offers advanced agent-building recipes · 2 sources tracked
Anthropic has released a collection of resources and tutorials, dubbed the Claude Cookbook, detailing how to build and deploy advanced AI agents using its Claude models and SDKs. These resources cover a range of applica…
-
Anthropic's Claude Agent SDK introduces plugins for streamlined tool integration
The Claude Agent SDK now supports plugins, allowing developers to bundle custom tools, subagents, hooks, and MCP servers into a single, portable directory. This simplifies agent development by eliminating the need to ma…
-
Anthropic's Claude Agent SDK Deployment Differs from AWS Lambda Architecture
Deploying an AI agent using the Claude Agent SDK requires a different architectural approach than traditional serverless functions like AWS Lambda. The SDK's components, including its shell, memory, and input handling, …
-
Claude Agent SDK uses hooks for deterministic AI agent guardrails
This article discusses the implementation of "hooks" within the Claude Agent SDK, a method for developers to establish deterministic guardrails for AI agents. These hooks allow for the blocking of destructive commands, …
-
OpenAI aids rare disease diagnosis; Anthropic pauses agent SDK billing change · 1 source tracked
AI news today covers a range of developments, including OpenAI's application of AI in rare disease diagnosis, Anthropic's decision to pause a controversial billing change for its Claude Agent SDK, and advancements in co…
-
Guide details 11 stages to build automated Claude AI assistant
This guide outlines an 11-stage process for building a self-running assistant using Anthropic's Claude models. It begins with establishing a baseline by manually solving a task in the chat interface to ensure prompt eff…
-
Anthropic's Claude Agent SDK: Exploring Five Design Patterns
This article explores five agent design patterns that can be implemented using Anthropic's Claude Agent SDK. It details how these patterns provide a framework for making design decisions when building AI agents with Claude.