This work proposes WebMCP-Phalanx, a dual-layer agent runtime architecture that provides a browser-native trust anchor that binds each tool to its registering principal through cryptographically protected capability credentials and propagates provenance labels throughout the tool lifecycle.
Abstract
The emerging W3C WebMCP proposal enables LLM agents to invoke tools exposed by web pages. In multi-party web environments, however, integrating agent execution into a browser security model centered on the Same-Origin Policy (SOP) leaves insufficient provenance and lifecycle guarantees for agent-accessible tools, creating three risks: subject-attribution spoofing, uncontrolled tool lifecycles, and semantic prompt injection. We propose WebMCP-Phalanx, a dual-layer agent runtime architecture. Its first layer provides a browser-native trust anchor that binds each tool to its registering principal through cryptographically protected capability credentials and propagates provenance labels throughout the tool lifecycle. Its second layer separates semantic inspection from privileged tool use. A Quarantine Agent (Q-LLM), without tool invocation authority, inspects tool metadata, outputs, and page-supplied content for prompt injection. Validated content is then forwarded to a Privileged Agent (P-LLM) for execution, while the Q-LLM's internal state remains hidden from page scripts. Empirical evaluation shows that the browser-native ownership mechanism reduces revocation and overwrite attack success from 100\% to 0\%. The dual-agent runtime blocks all 80 prompt-injection attempts embedded in tool descriptions and limits tool-return attacks to 2 successful cases out of 80. Across experiments, task utility remains statistically indistinguishable from the no-attack baseline. Under a white-box adaptive attacker, however, description-based filtering can be bypassed through malicious tool names invoked before inspection. This finding motivates a call-timing gate that delays tool invocation until all agent-visible tool metadata has been validated.
Agentic large language model (LLM) systems gain much of their practical value from tool calling, the capacity to invoke external functions such as web searches, database lookups, and application programming interface (API) requests during multi-step reasoning. Deploying such agents inside Trusted Execution Environments (TEEs) creates a structural tension: the confidential virtual machine (CVM) that protects user data must remain isolated from the host infrastructure, yet the agent must reach beyond the enclave boundary to be useful. The Model Context Protocol (MCP), which is now the main open standard for connecting LLM applications to external tools and data sources, was not designed with TEE constraints in mind, leaving three critical incompatibilities unresolved: transport mechanisms that expose user-derived parameters to untrusted hosts, dynamic capability discovery that violates pre-deployment transparency requirements, and authentication models misaligned with non-targetability guarantees. This paper presents Confidential MCP (C-MCP), a set of backward-compatible extensions to MCP that enable standardized, auditable tool calling within and across TEE boundaries. C-MCP introduces a three-zone enclave-partitioned server topology, a programmable Anonymization Transform Layer (ATL) with formal parameter classification and entropy bounds, and Attested Egress Policies (AEPs) that extend behavioral transparency from static binary attestation to constraints on verifiable runtime tool invocation. We analyze open-source LLM deployment challenges, including tool-calling information minimality, TEE inference overhead accumulation across agentic reasoning steps, and model supply chain integrity, and present concrete domain case studies in healthcare, legal practice, and financial services.
Ankur Aggarwal· International journal of com...· 0 citations
The results show that TraceGrant provides a unified governance layer that connects trusted user intent, runtime evidence, concrete tool execution, and verified task completion.
Bohao Liao, Jing-Chao Wang, Qipeng Song et al.· 0 citations
Large language model (LLM)-based agents are evolving into agents for computer use (ACUs) that read files, invoke applications, communicate over networks, and operate graphical interfaces, moving the effective security boundary from model inputs and outputs to autonomous actions that alter endpoint state. Conventional identity and access controls remain applicable and necessary, but they are authorized based on identity, resource, and network policy rather than on the semantic scope of the active task or the provenance of the instruction that triggered an action. This paper presents ARES (Agent Resource Enforcement and Security), an action-centric framework that inserts enforceable authorization between agent-generated tool calls and protected resources, combining a Resource Proxy Layer for interception, a Behavioral Guardrail Engine for task- and context-aware authorization, and a Multi-Agent Trust Boundary Manager for provenance and taint propagation. We implement ARES-lite and evaluate it against file exfiltration through indirect prompt injection, prompt infection propagation, and internal-network access abuse. Under the controlled replay-based evaluation, integrated ARES-lite reduced the observed attack-success rate from 100% under the permissive baseline to 0% across the three evaluated scenarios, while preserving the predefined handling of eight benign and ambiguous tasks with no false positives. A closed-loop evaluation in which the agent re-plans after an intervention, an adversarial boundary-case suite covering path aliasing, address encoding, look-alike destinations, and taint laundering, and a comparison against representative prompt-filtering and tool-allowlist defenses further characterize the enforcement path; a model and temperature sweep show that baseline exposure varies with model capability, whereas the mediated outcome does not.
Changhee Kim, Seong-je Cho· Electronics· 0 citations
Autonomous LLM agents can turn untrusted content into effectful actions such as payments and permission changes. If the same process interprets this content and controls a reusable signing credential, prompt injection can cross the judgment boundary and reach execution authority. We present KITA, a review-to-authorization architecture that keeps the user's personal secret signing key and every threshold signing-key share outside all LLM processes. Under threshold signature unforgeability and our system assumptions, compromising the proposer and fewer than t reviewer-signer domains cannot produce a valid authorization for a new action without signing contributions from t distinct domains. Thus, any such authorization includes a share from an uncompromised domain, bound to the canonical action and released only after authenticated reviewer approval. This establishes execution-bound authorization integrity. We implement the complete reviewer-to-executor path with a structured-output LLM adapter and threshold BLS. Six system tests validate quorum gating and message binding at this interface, while cryptographic microbenchmarks measure the online signing path and its scaling behavior.
Commercial LLM APIs advertise a specific foundation model, but the served backbone may be silently substituted, quantized, or wrapped, for example to save deployment costs. All existing audits decide backbone identity from the text-output channel, which is structurally fragile for agentic APIs because modern serving stacks (OpenAI, Anthropic, Gemini, Cloudflare Workers AI, LangGraph) discard text and expose only structured actions when the model calls a tool, and provider-injected system prompts can distort text distributions enough that text-channel tests falsely accuse honest providers of substituting the claimed model. We observe that recent agentic post-training internalizes tool-use directly into the weights, opening a new audit channel that the serving stack still exposes and that is largely invariant to deployment context. We introduce Agentic Provenance (AgentProv), the first action-based identity audit for agentic LLM APIs: AgentProv fingerprints a deployed model through its categorical tool-call distribution and decides identity via an MMD permutation test. AgentProv catches every substituted model (100% on 630 evaluated checkpoint pairs), while holding the false-positive rate under system-prompt injection at 7% (vs. 67% for MET and 53% for RUT). On third-party API endpoints, AgentProv's disagreements with MET are consistent with an independent token-count side-channel that detects provider-injected system prompts.
Xun Wang, Bihe Zhao, Michael Backes et al.· 0 citations
A comprehensive framework based on Tool-Flow Taint Analysis designed to mitigate data exfiltration in Large Language Model agents is introduced, providing a critical foundation for securing next-generation autonomous agents against sophisticated data-stealing attacks in enterprise environments.
Chun Tian, Hiu-Tung Li, Michelle Yu· Journal of innovative resear...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
