SkillsMetric: Mapping the Detection Boundary of Static Analysis for Malicious Agent Skills

• URL: http://arxiv.org/abs/2608.08468v1
• Date: Sun, 09 Aug 2026 04:19:10 GMT
• Title: SkillsMetric: Mapping the Detection Boundary of Static Analysis for Malicious Agent Skills
• Abstract summary: We present textscSkillsMetric, a five-stage static analysis framework that scores skill packages along pattern density, statistical anomaly, dataflow taint, import anomaly, and capability mismatch dimensions.<n>We construct an adversarial evaluation dataset of 2266 skills spanning 16attack types across code-level, system-level, and semantic-level threats, and evaluate on the full SkillMD-138K corpus.<n>Our framework achieves an AUC of 0.93 and 5-fold cross-validated F1 of 73.4%$pm$0.5%, with strong detection of data ex
• Score: 12.451393173022344
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Agent Skills---structured packages of instructions and scripts that augment LLM-based agents---are rapidly proliferating, yet their security properties remain under-explored. We present \textsc{SkillsMetric}, a five-stage static analysis framework that scores skill packages along pattern density, statistical anomaly, dataflow taint, import anomaly, and capability mismatch dimensions. We construct an adversarial evaluation dataset of 2{,}266 skills spanning 16~attack types across code-level, system-level, and semantic-level threats, and evaluate on the full SkillMD-138K corpus. Our framework achieves an AUC of 0.93 and 5-fold cross-validated F1 of 73.4\%$\pm$0.5\%, with strong detection of data exfiltration (93\%) and steganographic payloads (93\%). Crucially, we identify fundamental blind spots: \emph{host destruction} attacks using common shell commands evade all five stages (0\% detection), and \emph{prompt injection} via natural-language manipulation achieves only 42\% detection. These findings establish that static analysis alone is insufficient for skill security, motivating defense-in-depth architectures that combine fast static pre-screening with semantic review.

Related papers

• Behavioral Integrity Verification for AI Agent Skills [9.127363793428119]
  We formalize this as the behavioral integrity verification (BIV) problem.<n>On 49,943 skills from the OpenClaw registry, the deviation taxonomy reveals a pervasive description-implementation gap.<n>On a 906-skill malicious-skill detection benchmark, BIV reaches an F1 of 0.946, outperforming state-of-the-art rule-based and single-pass LLM baselines.
  arXiv  Detail & Related papers  (2026-05-12T08:41:09Z)
• You Told Me to Do It: Measuring Instructional Text-induced Private Data Leakage in LLM Agents [9.719776777345364]
  High-privilege LLM agents that autonomously process external documentation are increasingly trusted to automate tasks.<n>These agents are granted terminal access, control, and outbound network connectivity with minimal security oversight.<n>We measure a fundamental vulnerability in this trust model, which we term the emphTrusted Executor Dilemma<n>This vulnerability is a structural consequence of the instruction-following design paradigm, not an implementation bug.
  arXiv  Detail & Related papers  (2026-03-12T12:35:46Z)
• When Detectors Forget Forensics: Blocking Semantic Shortcuts for Generalizable AI-Generated Image Detection [48.75021863792367]
  Vision Foundation Models (VFMs, emphe.g., CLIP) often struggle to generalize to images created using unseen generation pipelines.<n>We propose textbfGeometric Semantic Decoupling (GSD), a parameter-free module that explicitly removes semantic components from learned representations.<n>GSD estimates semantic directions from batch-wise statistics and projects them out via a geometric constraint, forcing the artifact detector to rely on semantic-invariant forensic evidence.
  arXiv  Detail & Related papers  (2026-03-10T06:16:35Z)
• When Benchmarks Lie: Evaluating Malicious Prompt Classifiers Under True Distribution Shift [0.0]
  We present a comprehensive analysis using a benchmark of 18 datasets spanning harmful requests, jailbreaks, indirect prompt injections, and extraction attacks.<n>We propose Leave-One-Dataset-Out (LODO) evaluation to measure true out-of-distribution generalization.
  arXiv  Detail & Related papers  (2026-02-15T14:21:43Z)
• Prompt Injection Attacks on Agentic Coding Assistants: A Systematic Analysis of Vulnerabilities in Skills, Tools, and Protocol Ecosystems [0.0]
  We present a comprehensive analysis of prompt injection attacks targeting agentic coding assistants.<n>Our meta-analysis synthesizes findings from 78 recent studies.<n>Our findings indicate that the security community must treat prompt injection as a first-class vulnerability class.
  arXiv  Detail & Related papers  (2026-01-24T18:39:21Z)
• Multi-Agent Taint Specification Extraction for Vulnerability Detection [49.27772068704498]
  Static Application Security Testing (SAST) tools using taint analysis are widely viewed as providing higher-quality vulnerability detection results.<n>We present SemTaint, a multi-agent system that strategically combines the semantic understanding of Large Language Models (LLMs) with traditional static program analysis.<n>We integrate SemTaint with CodeQL, a state-of-the-art SAST tool, and demonstrate its effectiveness by detecting 106 of 162 vulnerabilities previously undetectable by CodeQL.
  arXiv  Detail & Related papers  (2026-01-15T21:31:51Z)
• Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale [26.757365536859453]
  The rise of AI agent frameworks has introduced agent skills, modular packages containing instructions and executable code that dynamically extend agent capabilities.<n>While this architecture enables powerful customization, skills execute with implicit trust and minimal vetting, creating a significant yet uncharacterized attack surface.<n>We conduct the first large-scale empirical security analysis of this emerging ecosystem, collecting 42,447 skills from two major marketplaces.
  arXiv  Detail & Related papers  (2026-01-15T12:31:52Z)
• AutoMalDesc: Large-Scale Script Analysis for Cyber Threat Research [81.04845910798387]
  Generating natural language explanations for threat detections remains an open problem in cybersecurity research.<n>We present AutoMalDesc, an automated static analysis summarization framework that operates independently at scale.<n>We publish our complete dataset of more than 100K script samples, including annotated seed (0.9K) datasets, along with our methodology and evaluation framework.
  arXiv  Detail & Related papers  (2025-11-17T13:05:25Z)
• VulAgent: Hypothesis-Validation based Multi-Agent Vulnerability Detection [55.957275374847484]
  VulAgent is a multi-agent vulnerability detection framework based on hypothesis validation.<n>It implements a semantics-sensitive, multi-view detection pipeline, each aligned to a specific analysis perspective.<n>On average, VulAgent improves overall accuracy by 6.6%, increases the correct identification rate of vulnerable--fixed code pairs by up to 450%, and reduces the false positive rate by about 36%.
  arXiv  Detail & Related papers  (2025-09-15T02:25:38Z)
• ASSERT: Automated Safety Scenario Red Teaming for Evaluating the Robustness of Large Language Models [65.79770974145983]
  ASSERT, Automated Safety Scenario Red Teaming, consists of three methods -- semantically aligned augmentation, target bootstrapping, and adversarial knowledge injection. We partition our prompts into four safety domains for a fine-grained analysis of how the domain affects model performance. We find statistically significant performance differences of up to 11% in absolute classification accuracy among semantically related scenarios and error rates of up to 19% absolute error in zero-shot adversarial settings.
  arXiv  Detail & Related papers  (2023-10-14T17:10:28Z)
• How Robust are Randomized Smoothing based Defenses to Data Poisoning? [66.80663779176979]
  We present a previously unrecognized threat to robust machine learning models that highlights the importance of training-data quality. We propose a novel bilevel optimization-based data poisoning attack that degrades the robustness guarantees of certifiably robust classifiers. Our attack is effective even when the victim trains the models from scratch using state-of-the-art robust training methods.
  arXiv  Detail & Related papers  (2020-12-02T15:30:21Z)

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.