Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture

• URL: http://arxiv.org/abs/2607.22445v1
• Date: Fri, 24 Jul 2026 16:08:03 GMT
• Title: Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture
• Abstract summary: We argue that capability scoping must follow a dynamic least-privilege principle and be treated as a prevention mechanism before a detection.<n>A credential that does not exist in an agent's context cannot be misused regardless of the agent's reasoning.<n>We contribute a synthetic dataset of 600 enterprise task prompts grounded in a multi-department company policy.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Enterprise AI agents are typically granted static credential sets at configuration time, holding every tool the role might need for every task they perform. This persistent over-privilege expands the attack surface. We argue that capability scoping must follow a dynamic least-privilege principle and be treated as a prevention mechanism before a detection one. A credential that does not exist in an agent's context cannot be misused regardless of the agent's reasoning or evasion sophistication. We outline a three-source architecture instantiating this principle: role-based ceilings, a task-context classifier, and policy-derived combination prohibitions creating a layered proactive defense against LLM agent misalignment and misuse cases. The architecture supports both enforcing and observe-only deployment; the latter records agent permission requests inconsistent with task context, producing a behavioral signal usable in misalignment research. As a first step toward evaluating this architecture, we contribute a synthetic dataset of 600 enterprise task prompts grounded in a multi-department company policy, labeled with minimum required permissions across a 15-permission tool-based taxonomy that maps directly to deployable credentials or enforceable guardrails. The dataset is constructed via a two-pass pipeline that separates prompt generation from permission labeling to avoid circularity, and is validated against a 60-record/688 decisions human-reviewed sample (Cohen's $κ= 0.917$ pre-review and $κ= 0.967$ post-review). Iterating between dataset and policy reduced ceiling violations from 46 to 3, a 93% reduction. This shows that synthetic prompt generation can drive policy refinement when the two are developed together. The dataset, environment specification, and generation pipeline are released to support evaluation of dynamic scoping mechanisms.

Related papers

• Authorization Architectures for Tool-Using AI Agents [0.0]
  Trustworthy human-AI systems require that every consequential agent action be traceable to a human principal.<n>This review introduces a principal hierarchy spanning human user, operator/deployer, orchestrator agent, sub-agent, and tool endpoint as an organizing framework.<n>We propose seven structural requirements, derive a four-layer reference architecture, apply the requirements to three deployable reference configurations, and identify runtime enforcement and aggregation bounds as the principal unresolved gaps.
  arXiv  Detail & Related papers  (2026-09-14T17:21:22Z)
• Formal Verification of Agentic Systems over Operational Data [59.98246281888422]
  Agentic systems driven by large language models (LLMs) are increasingly deployed in real-world where they act on persistent operational data.<n>Existing approaches do not provide such system-level guarantees.<n>We study here the verification of agentic systems comprising a single LLM and a tool orchestration harness over relational operational data.
  arXiv  Detail & Related papers  (2026-08-04T13:01:30Z)
• ScrambleToolBench: Agents Search Exhaustively Even When Their Own Map Points to the Next Step [74.83535434786145]
  We introduce ScrambleToolBench, an interactive terminal benchmark designed to isolate behavioral reasoning.<n>The benchmark requires agents to uncover hidden tool behaviors entirely through trial-and-error interaction.<n>Our evaluation of state-of-the-art language models reveals that successful initial discovery does not translate into robust adaptation.
  arXiv  Detail & Related papers  (2026-08-03T15:07:46Z)
• Agentic Permissions Policy Algebra for Taint Confinement in LLM Agents [0.0]
  Traditional taint tracking permanently taints an agent's context upon reading unvetted data.<n>We present APPA, an IFC framework that resolves this usability bottleneck through engine-managed context branching.<n>We evaluate APPA on a multi-turn tool-chaining benchmark across four models.
  arXiv  Detail & Related papers  (2026-07-27T16:19:45Z)
• PRIMA: Operational Patterns for Resilient Multi-Agent Research with Verifiable Identity and Convergent Feedback [0.0]
  PRIMA operates as coordinated multi-agent research systems over multi-hour runs.<n>Main contributions are three operational patterns for surviving failure modes.<n>A Graph Isomorphism case study grounds the architectural claims in a generated artifact.
  arXiv  Detail & Related papers  (2026-05-23T23:27:46Z)
• Securing LLM Agents Need Intent-to-Execution Integrity [49.490963596514185]
  We argue that securing LLM agents requires defining an end-to-end correctness property that specifies when an agent's execution faithfully reflects the user's intent.<n>We observe that LLM agents are structurally analogous to compilers, where security violations correspond to mis-executions that do not preserve user intent.<n>We derive four integrity properties that must hold simultaneously: emphTool Integrity, emphInstruction Integrity, emphJudgment Integrity, and emphData Flow Integrity.
  arXiv  Detail & Related papers  (2026-05-16T12:53:31Z)
• PRISM: : Planning and Reasoning with Intent in Simulated Embodied Environments [59.07829883257003]
  Built on five multi-room apartments, PRISM structures 300 human-verified tasks into three capability tiers.<n>PRISM exposes an agent-agnostic executable action API that allows arbitrary agents to be evaluated end-to-end.
  arXiv  Detail & Related papers  (2026-05-12T04:59:47Z)
• Securing the Agent: Vendor-Neutral, Multitenant Enterprise Retrieval and Tool Use [0.0]
  We introduce a layered isolation architecture combining policy-aware ingestion, retrieval-time gating, and shared inference.<n>This approach centralizes security-critical operations--tool execution authorization, state isolation, and policy enforcement--on the server.<n>We evaluate it empirically and show that ABAC gating eliminates cross-tenant leakage while introducing negligible overhead.
  arXiv  Detail & Related papers  (2026-05-06T17:59:21Z)
• Semantic Intent Fragmentation: A Single-Shot Compositional Attack on Multi-Agent AI Pipelines [0.9772267314090434]
  We introduce Semantic Intent Fragmentation (SIF) as an attack class against LLM orchestration systems.<n>SIF exploits four mechanisms: bulk scope, silent data exfiltration, embedded trigger deployment, and quasi-identifier aggregation.<n>A GPT-20B orchestrator produces policy-violating plans in 71% of cases while every subtask appears benign.
  arXiv  Detail & Related papers  (2026-04-08T18:19:03Z)
• Dynamic analysis enhances issue resolution [53.50448142467294]
  DAIRA (Dynamic Analysis-enhanced Issue Resolution Agent) is an automated repair framework that embeds dynamic analysis into the agent's reasoning cycle.<n> Driven by a Test Tracing-Driven methodology, DAIRA utilizes lightweight monitors to extract critical runtime data.<n>When powered by Gemini 3 Flash Preview, DAIRA establishes a new state-of-the-art (SOTA) performance, achieving a 79.4% resolution rate on the SWE-bench Verified dataset.
  arXiv  Detail & Related papers  (2026-03-23T14:48:54Z)
• Agentic Peer-to-Peer Networks: From Content Distribution to Capability and Action Sharing [10.47562113256175]
  This article outlines the networking foundations needed to make such collaboration practical.<n>We propose a plane-based reference architecture that decouples connectivity/identity, semantic discovery, and execution.<n>We also present a textittiered verification spectrum: Tier1 relies on reputation signals, Tier2 applies lightweight canary challenge-response with fallback selection, and Tier3 requires evidence packages such as signed tool receipts/traces.
  arXiv  Detail & Related papers  (2026-03-04T05:58:44Z)
• BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents [58.83028403414688]
  Large language model (LLM) agents execute tasks through multi-step workflow that combine planning, memory, and tool use.<n>Backdoor triggers injected into specific stages of an agent workflow can persist through multiple intermediate states and adversely influence downstream outputs.<n>We propose textbfBackdoorAgent, a modular and stage-aware framework that provides a unified agent-centric view of backdoor threats in LLM agents.
  arXiv  Detail & Related papers  (2026-01-08T03:49:39Z)
• Analyzing and Internalizing Complex Policy Documents for LLM Agents [53.14898416858099]
  Large Language Model (LLM)-based agentic systems rely on in-context policy documents encoding diverse business rules.<n>This motivates developing internalization methods that embed policy documents into model priors while preserving performance.<n>We introduce CC-Gen, an agentic benchmark generator with Controllable Complexity across four levels.
  arXiv  Detail & Related papers  (2025-10-13T16:30:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.