相关论文: ChainCaps: Composition-Safe Tool-Using Agents via …

Agentic AI systems built around large language models (LLMs) are moving away from closed, single-model frameworks and toward open ecosystems that connect a variety of agents, external tools, and resources. The Model Context Protocol (MCP)…

Agentic systems increasingly solve complex user requests by executing orchestrated workflows, where subtasks are assigned to specialized models or tools and coordinated according to their dependencies. While recent work improves agent…

Tool-using automation systems, from scripts and CI bots to agentic assistants, fail in recurring patterns. Common failures include unsafe side effects, invalid arguments, uncontrolled retries, and leakage of sensitive outputs. Many…

Driven by Large Language Models, the single-agent, multi-tool architecture has become a popular paradigm for autonomous agents. However, this architecture introduces a severe privacy risk, which we term Tools Orchestration Privacy Risk…

Large language model (LLM)-based AI agents extend LLM capabilities by enabling access to tools such as data sources, APIs, search engines, code sandboxes, and even other agents. While this empowers agents to perform complex tasks, LLMs may…

Designing effective agentic systems requires the seamless composition and integration of agents, tools, and models within dynamic and uncertain environments. Most existing methods rely on static, semantic retrieval approaches for tool or…

Personal AI agents like OpenClaw run with elevated privileges on users' local machines, where a single successful prompt injection can leak credentials, redirect financial transactions, or destroy files. This threat goes well beyond…

As LLMs advance into autonomous agents with tool-use capabilities, they introduce security challenges that extend beyond traditional content-based LLM safety concerns. This paper introduces Sequential Tool Attack Chaining (STAC), a novel…

OpenClaw-like agents offer substantial productivity benefits, yet they are insecure by default because they combine untrusted inputs, autonomous action, extensibility, and privileged system access within a single execution loop. We use…

Tool calling has emerged as a critical capability for AI agents. In contrast to conventional tool calling frameworks that rely on static, provider-specific tool definitions, the Model Context Protocol (MCP) offers a unified interface to…

Security mandates today are often in the form of checklists and are generally inflexible and slow to adapt to changing threats. This paper introduces an alternate approach called open mandates, which mandate that vendors must dedicate some…

The Model Context Protocol (MCP) is a recently proposed interoperability standard that unifies how AI agents connect with external tools and data sources. By defining a set of common client-server message exchange clauses, MCP replaces…

Coding agents often pass per-prompt safety review yet ship exploitable code when their tasks are decomposed into routine engineering tickets. The challenge is structural: existing safety alignment evaluates overt requests in isolation,…

Large Language Models produce a controllability gap in safety-critical engineering: even low rates of undetected constraint violations render a system undeployable. Current orchestration paradigms suffer from sycophantic compliance, context…

By providing a standardized interface for LLM agents to interact with external tools, the Model Context Protocol (MCP) is quickly becoming a cornerstone of the modern autonomous agent ecosystem. However, it creates novel attack surfaces due…

Model Context Protocol (MCP) is increasingly adopted for tool-integrated LLM agents, but its multi-layer design and third-party server ecosystem expand risks across tool metadata, untrusted outputs, cross-tool flows, multimodal inputs, and…

The Model Context Protocol (MCP) has become a common interface for connecting large language model (LLM) agents to external tools, but its reliance on stateless, eager schema injection imposes a hidden per-turn overhead the MCP Tax or Tools…

Multi-agent systems increasingly expose explicit workflow structure: agents, tools, tool-access rules, restrictions, and delegation paths. Existing evaluations rely largely on end-to-end task success, benchmark scores, final-response…

Tool-using LLM agents must act on untrusted webpages, emails, files, and API outputs while issuing privileged tool calls. Existing defenses often mediate trust at the granularity of an entire tool invocation, forcing a brittle choice in…

From software development to robot control, modern agentic systems decompose complex objectives into a sequence of subtasks and choose a set of specialized AI agents to complete them. We formalize agentic workflows as directed acyclic…