mlsanityData to 14 Oct 2026

SkillAtlas: An Attack Trace Library for Agent Skills

arXiv 2609.13353 ↗First posted 11 Sep 2026cs.CR cs.AIcited by 0

On the map: LLM agents and tool use → Agent threats, guardrails and governance

Abstract (arXiv)

Agent skills are reusable units for language-model agents, but their risks emerge through model decisions, user context, tool calls, and execution feedback rather than through stable signatures or a single sandbox run. Existing static, dynamic, and benchmark-style evaluations rarely preserve public evidence that can be inspected, searched, and reused. We present SkillAtlas, a hosted attack trace library that converts private agent-skill security report bundles into reviewed, redacted, and searchable public cases. The library contains 3,014 cases, 6,589 traces, 151,131 steps, 233 affected skills, and 8 risk categories; 42.5% of successful cases first become successful after a non-success initial round, and trajectory-grounded labels improve pre-execution guard accuracy to 0.770.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use. Builds on 24 of its 46 references that are in the map. First posted on arXiv 11 Sep 2026.

Builds on

24 of 46 references.

Closest ancestors · 8

Same direction, newest first

  1. 2026SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems8 Apr
  2. 2026SkillAttack: Automated Red Teaming of Agent Skills through Attack Path Refinement5 Apr
  3. 2026Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks23 Feb
  4. 2026SkillJect: Effectively Automating Skill-Based Prompt Injection for Skill-Enabled Agents15 Feb
  5. 2026Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale15 Jan
  6. 2025Agent Skills Enable a New Class of Realistic and Trivially Simple Prompt Injections30 Oct
  7. 2024AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents11 Oct
  8. 2023Identifying the Risks of LM Agents with an LM-Emulated Sandbox25 Sep

Other references · 16

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  3. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  4. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 3,104
  5. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,947
  6. 2023Jailbroken: How Does LLM Safety Training Fail?cited by 1,651
  7. 2023Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injectioncited by 1,443
  8. 2023Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversationscited by 1,207
  9. 2022Red Teaming Language Models with Language Modelscited by 1,026
  10. 2023NeMo Guardrails: A Toolkit for Controllable and Safe LLM Applications with Programmable Railscited by 365
  11. 2025MCPEval: Automatic MCP-based Deep Evaluation for AI Agent Modelscited by 36
  12. 2024UltraEval: A Lightweight Platform for Flexible and Comprehensive Evaluation for LLMscited by 24
4 more
  1. 2022A Comparison of Static, Dynamic, and Hybrid Analysis for Malware Detectioncited by 17
  2. 2024OpenEval: Benchmarking Chinese LLMs across Capability, Alignment and Safetycited by 9
  3. 2025ResearStudio: A Human-Intervenable Framework for Building Controllable Deep-Research Agentscited by 3
  4. 2024BotEval: Facilitating Interactive Human Evaluationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Agent Skill Security: Threat Models, Attacks, Defenses, and Evaluation
  2. 2026SkillSafetyBench: Evaluating Agent Safety under Skill-Facing Attack Surfaces
  3. 2026SkillSecurer: Detecting and Patching Prompt-Injection Vulnerabilities in AI Agent Skills
  4. 2026SkillsMetric: Mapping the Detection Boundary of Static Analysis for Malicious Agent Skills
  5. 2026MaliciousSkillBench: A Comprehensive Benchmark for Malicious Agent Skill Detection
  6. 2026SkillSentry: Adaptive Honey Worlds for Dynamic Safety Testing of Agent Skills
  7. 2026SkillMutator: Benchmarking and Defending Language-and-Code Cross-modal Attacks on LLM Agent Skills
  8. 2026SkillHarm: Lifecycle-Aware Skill-Based Attacks via Automated Construction