Back to list
LangChain LangSmith Fleet Introduces Two Distinct Agent Authorization Models: Assistants and Claws
Product LaunchLangChainAI AgentsCybersecurity

LangChain LangSmith Fleet Introduces Two Distinct Agent Authorization Models: Assistants and Claws

LangChain has officially introduced two specialized types of agent authorization within its LangSmith Fleet platform: Assistants and Claws. This update addresses the critical need for flexible credential management in AI agent deployment. The 'Assistants' model is designed to operate using the end user's own credentials, ensuring personalized and user-specific access. In contrast, the 'Claws' model utilizes a fixed set of credentials, providing a standardized approach for agent operations. These two distinct paths offer developers more granular control over how agents interact with protected resources and manage security permissions, marking a significant step in the evolution of agentic workflows and secure integration within the LangChain ecosystem.

LangChain
Share

Key Takeaways

  • LangSmith Fleet has launched two new authorization frameworks for AI agents.
  • Assistants utilize the specific credentials of the end user for authentication.
  • Claws operate using a pre-defined, fixed set of credentials.
  • The update provides developers with flexible options for managing security and access control.

In-Depth Analysis

The Assistants Model: User-Centric Authorization

The first authorization type introduced by LangSmith Fleet is the Assistants model. This approach is fundamentally built around the end user's identity. By using the end user's own credentials, Assistants can perform tasks and access data that are specifically permitted for that individual. This ensures that the agent acts as a direct extension of the user, maintaining the same security boundaries and permissions that the user would have when interacting with a system manually. This model is particularly useful for applications where personalized data access and individual accountability are paramount.

The Claws Model: Fixed Credential Management

The second authorization type is known as Claws. Unlike the Assistants model, Claws do not rely on varying user identities; instead, they function using a fixed set of credentials. This method is ideal for scenarios where an agent needs to perform background tasks, access shared resources, or operate within a controlled environment where the identity of the individual user is less relevant than the identity of the service itself. By utilizing a consistent set of credentials, Claws simplify the management of service-level permissions and provide a stable framework for automated agent actions.

Industry Impact

The introduction of these two authorization types by LangChain represents a significant advancement in the professionalization of AI agent deployment. By distinguishing between user-owned credentials (Assistants) and fixed credentials (Claws), LangChain is addressing a core challenge in AI security: how to grant agents the power to act while maintaining strict access controls. This development allows for more sophisticated enterprise integrations, as organizations can now choose the authorization method that best fits their specific security protocols and operational requirements. It sets a precedent for how agentic platforms should handle the delicate balance between autonomy and security.

Frequently Asked Questions

Question: What is the main difference between Assistants and Claws in LangSmith Fleet?

Assistants use the credentials belonging to the end user, whereas Claws use a fixed set of credentials regardless of the end user.

Question: Which authorization type should be used for personalized user tasks?

The Assistants model is designed for personalized tasks as it operates under the end user's own credentials.

Question: What is the purpose of the Claws authorization type?

Claws are intended for operations that require a stable, fixed set of credentials, making them suitable for service-level tasks or shared resource access.

Read original article

Related News

How to Use LangSmith for Fine-Tuning Open-Source LLMs Like LLaMA2 and GPT-3.5
Product Launch

How to Use LangSmith for Fine-Tuning Open-Source LLMs Like LLaMA2 and GPT-3.5

LangChain has introduced a comprehensive guide detailing how LangSmith supports the fine-tuning and evaluation of Large Language Models (LLMs). The update focuses on enhancing dataset management, providing developers with the tools necessary to refine model performance effectively. The guide specifically highlights practical examples for fine-tuning both open-source models like LLaMA2 and proprietary models such as GPT-3.5. By integrating LangSmith into the fine-tuning workflow, users can better manage datasets and evaluate the outcomes of their training processes. This development marks a significant step in providing structured support for the lifecycle of LLM development, from data preparation to final model evaluation.

Instagram Launches First Draft Feature to Automatically Trim Reels and Highlight Key Video Moments
Product Launch

Instagram Launches First Draft Feature to Automatically Trim Reels and Highlight Key Video Moments

Instagram has introduced a new feature called "First Draft" to its Reels platform, aimed at streamlining the video editing process for creators. The tool automatically trims video clips to focus on the most important highlights, providing a foundational "starting point" for further customization. Currently rolling out to the Instagram iPhone app, First Draft is designed to reduce the manual effort required to edit raw footage into engaging short-form content. By identifying key moments automatically, the feature allows users to quickly transition from capturing footage to the final creative stages of editing. This update reflects Instagram's commitment to lowering the barrier to entry for video creation by offering automated tools that assist in the initial assembly of Reels.

Inside IBM Granite 4.2: A Technical Deep Dive into the New Era of Open-Source Reasoning and Agentic LLMs
Product Launch

Inside IBM Granite 4.2: A Technical Deep Dive into the New Era of Open-Source Reasoning and Agentic LLMs

IBM has officially unveiled Granite 4.2, a groundbreaking family of dense, decoder-only large language models (LLMs) designed specifically for enterprise-grade reasoning and agentic workflows. Released in 3B, 8B, and 30B parameter sizes under the Apache 2.0 license, these models represent a significant leap in open-source AI capabilities. Granite 4.2 is trained on approximately 15 trillion tokens using a sophisticated five-phase strategy that extends its context window to 512K tokens. A key innovation is the introduction of native reasoning—a switchable "thinking" mode that allows the models to perform step-by-step chain-of-thought deliberation. By integrating agentic reinforcement learning (RL) within real-world sandboxed environments like OpenHands and terminal interfaces, IBM has optimized the 8B and 30B versions for complex software engineering and tool-calling tasks, setting a new benchmark for open, transparent, and high-performance AI agents.