Claude Blog 采集 (2026-06-04)¶
共采集 5 篇文章
📋 文章索引¶
- Using LLMs to secure source code - May 27, 2026 (评分: 9.0)
- Claude can now search the web - Mar 20, 2025 (评分: 9.0)
- Introducing web search on the Anthropic API - May 07, 2025 (评分: 9.0)
- What is Model Context Protocol? Connect AI to your world - Oct 31, 2025 (评分: 9.0)
- Workspaces in the Anthropic API Console - Sep 10, 2024 (评分: 9.0)
Using LLMs to secure source code¶
来源: Claude Blog 发布日期: May 27, 2026 采集时间: 2026-06-04 价值评分: 9.0/10 正文字数: ~2263 字符
摘要¶
We share best practices for how you can work with Claude Opus to build a threat model, discover vulnerabilities in your codebase, then verify, triage, and patch them.
正文内容¶
We share best practices for how you can work with Claude Opus to build a threat model, discover vulnerabilities in your codebase, then verify, triage, and patch them.
Model capabilities are advancing quickly, and unevenly. We’ve been working with security teams to find and fix vulnerabilities in their own code and open source software, and the work has given us a better understanding of how to use models to secure source code. Our primary takeaway: discovery is now straightforward to parallelize, and the bottleneck has shifted to verification, triage, and patching . To give some indication of this discrepancy, as part of our own scanning of open source software, as of May 22, 2026, we had disclosed 1,596 vulnerabilities. To our knowledge, 97 of these have been patched. This guide walks through how you can work with Claude Opus to build a threat model, discover vulnerabilities in your codebase, then verify, triage, and patch them. While we don’t have all the answers, we’ll share how teams have scaled discovery and what’s helped in the later stages. Get started today with the accompanying repo which includes skills for interactive workflows and a demo harness for autonomous scanning; we’ll call out the skill that implements each step as you read. The find-and-fix loop Teams finding and fixing the most vulnerabilities converged on a variation of existing best practices. We’ve distilled them into a sequence of six steps: Threat model: Decide what counts as a vulnerability before you start scanning. Sandbox: Build a sandbox environment to isolate agents and prove exploits. Discovery: Have models look for vulnerabilities in your source code. Verification: Independently confirm which findings are actually exploitable. Triage: Deduplicate findings, assign severity, and prioritize what needs fixing. Patching: Apply the fix, confirm the vulnerability is nullified, and search for variants.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Claude can now search the web¶
来源: Claude Blog 发布日期: Mar 20, 2025 采集时间: 2026-06-04 价值评分: 9.0/10 正文字数: ~2257 字符
摘要¶
Claude can now search the internet to provide up-to-date, cited responses on all plans worldwide. Get real-time insights for research, market analysis, and more.
正文内容¶
Claude can now search the web to deliver up-to-date, cited information in conversational responses across all plans.
Update: Web search is now available globally on all Claude plans. (May 27, 2025) You can now use Claude to search the internet to provide more up-to-date and relevant responses. With web search, Claude has access to the latest events and information, boosting its accuracy on tasks that benefit from the most recent data. When Claude incorporates information from the web into its responses, it provides direct citations so you can easily fact check sources. Instead of finding search results yourself, Claude processes and delivers relevant sources in a conversational format. This enhancement expands Claude's extensive knowledge base with real-time insights, providing answers based on more current information. Popular ways to use Claude with web search: Sales teams can transform account planning and drive higher win rates through informed conversations with prospects by analyzing industry trends to learn key initiatives and pain points. Financial analysts can assess current market data, earnings reports, and industry trends to make better investment decisions and inform financial model assumptions. Researchers can build stronger grant proposals and literature reviews by searching across primary sources on the web, spotting emerging trends and identifying gaps in the current literature. Shoppers can compare product features, prices, and reviews across multiple sources to make more informed purchase decisions. Getting started Web search is available now in feature preview for all paid Claude users in the United States. Support for users on our free plan and more countries is coming soon. To get started, toggle on web search in your profile settings and start a conversation with Claude 3.7 Sonnet. When applicable, Claude will search the web to inform its response.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Introducing web search on the Anthropic API¶
来源: Claude Blog 发布日期: May 07, 2025 采集时间: 2026-06-04 价值评分: 9.0/10 正文字数: ~1119 字符
摘要¶
Claude can now search the web through the API, giving users access to real-time information with citations for building up-to-date AI applications. Unknown block type "latestUpdates", specify a component for it in the `components.
正文内容¶
Claude can now search the web through the API, giving users access to real-time information with citations for building up-to-date AI applications.
Update: You can now add the web fetch tool to your requests and Claude will fetch and analyze content from any webpage URL. September 10, 2025 Today, we're introducing web search on the Anthropic API—a new tool that gives Claude access to current information from across the web. With web search enabled, developers can build Claude-powered applications and agents that deliver up-to-date insights. Power AI agents with the latest information from the web Developers can now augment Claude’s comprehensive knowledge with current, real-world data by enabling the web search tool when making requests to the Messages API.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
What is Model Context Protocol? Connect AI to your world¶
来源: Claude Blog 发布日期: Oct 31, 2025 采集时间: 2026-06-04 价值评分: 9.0/10 正文字数: ~3062 字符
摘要¶
Connect AI assistants to your tools without custom integrations using Model Context Protocol. AI models are only as good as the context provided to them. AI assistants like Claude can answer questions and perform an impressive range of tasks, but if they can't access the data or tools they need,...
正文内容¶
Connect AI assistants to your tools without custom integrations using Model Context Protocol.
AI models are only as good as the context provided to them. AI assistants like Claude can answer questions and perform an impressive range of tasks, but if they can't access the data or tools they need, they're limited in what they can do for you. You typically solve this by copying and pasting context from one tab to another, whether it's editing a document in Google Drive, replying to a thread in Slack, or updating code in an IDE. This process is slow, manual, and risks leaving out important context. The Model Context Protocol (MCP) offers a solution that is open and widely available across all AI apps and assistants. In this article, you'll learn what MCP is, how it works and why it matters, and who it's for. You'll see examples of MCP in action and understand how you can start using or building with MCP today. What is the Model Context Protocol (MCP)? The Model Context Protocol is an open standard that defines how LLMs communicate with external systems. Think of MCP as USB-C for LLMs . Just as USB-C provides a universal connector for your phone, laptop, and other devices, MCP provides a universal format for LLMs to connect with external systems. Before USB-C, every electronic gadget had its own cable: Lightning for iPhone, micro-USB for Android, proprietary connectors for cameras. As more devices adopted USB-C, connectivity became seamless across the ecosystem. MCP brings this same simplicity to AI integrations. Before MCP, every application and database required custom code to connect with LLMs. Google Drive needed its own integration, Slack needed another, Figma yet another. Now, MCP provides a single, standardized format for connecting these tools to Claude and other AI applications.
No. MCP is an open-source protocol. While Claude pioneered adoption of MCP, other AI providers have now adopted the same protocol, allowing anyone to connect to the same ecosystem of MCP servers.
Not for using connectors . Browse, install, authenticate. That's it. Building custom MCP servers requires TypeScript or Python knowledge, but the growing connector library covers most mainstream tools.
Each server requests specific permissions to allow Claude to access it. You can approve or deny access and revoke permissions at any time.
MCP uses efficient protocols. Stdio transport for local servers provides minimal overhead. Server-sent events (SSE) and Streamable HTTP for remote servers maintain persistent connections. Response streaming prevents timeouts on large data operations. The protocol supports pagination, filtering, and partial responses to handle large datasets efficiently.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集
Workspaces in the Anthropic API Console¶
来源: Claude Blog 发布日期: Sep 10, 2024 采集时间: 2026-06-04 价值评分: 9.0/10 正文字数: ~979 字符
摘要¶
Workspaces make it easier for developers to manage multiple Claude deployments at once. We're introducing Workspaces in the Anthropic API Console to help developers efficiently manage multiple Claude deployments.
正文内容¶
Workspaces make it easier for developers to manage multiple Claude deployments at once.
We're introducing Workspaces in the Anthropic API Console to help developers efficiently manage multiple Claude deployments. Workspaces are unique environments that enable you to organize resources, streamline access controls, and set custom spend and rate limits on a more granular level. Managing multiple deployments For developers using Claude across different environments—like development, staging, and production—and different use cases, Workspaces provide an abstraction layer for your overall organization and individual API keys.
Explore more product news and best practices for teams building with Claude.
Transform how your organization operates with Claude
Product updates, how-tos, community spotlights, and more. Delivered monthly to your inbox.
Please provide your email address if you'd like to receive our monthly developer newsletter. You can unsubscribe at any time.
采集自 Claude Blog，由 collect_claude_blog.py 自动采集