Search papers, labs, and topics across Lattice.
The paper introduces the Scheduling-Structural-Logical (SSL) representation, a structured format for agent skills that disentangles scheduling, execution structure, and logic-level action evidence, addressing the limitations of text-heavy skill representations in LLM agents. They instantiate SSL using an LLM-based normalizer and evaluate it on Skill Discovery and Risk Assessment tasks. Results show that SSL outperforms text-only baselines, improving MRR from 0.573 to 0.707 in Skill Discovery and macro F1 from 0.744 to 0.787 in Risk Assessment, demonstrating the benefits of explicit structure for skill management.
LLM agents can better discover and assess risks of skills when those skills are represented in a structured format that explicitly represents scheduling, execution structure, and logic, rather than relying on unstructured text.
LLM agents increasingly rely on reusable skills, capability packages that combine instructions, control flow, constraints, and tool calls. In most current agent systems, however, skills are still represented by text-heavy artifacts, including SKILL{.}md-style documents and structured records whose machine-usable evidence remains embedded largely in natural-language descriptions. This poses a challenge for skill-centered agent systems: managing skill collections and using skills to support agent both require reasoning over invocation interfaces, execution structure, and concrete side effects that are often entangled in a single textual surface. An explicit representation of skill knowledge may therefore help make these artifacts easier for machines to acquire and leverage. Drawing on Memory Organization Packets, Script Theory, and Conceptual Dependency from Schank and Abelson's classical work on linguistic knowledge representation, we introduce what is, to our knowledge, the first structured representation for agent skill artifacts that disentangles skill-level scheduling signals, scene-level execution structure, and logic-level action and resource-use evidence: the Scheduling-Structural-Logical (SSL) representation. We instantiate SSL with an LLM-based normalizer and evaluate it on a corpus of skills in two tasks, Skill Discovery and Risk Assessment, and superiorly outperform the text-only baselines: in Skill Discovery, SSL improves MRR from 0.573 to 0.707; in Risk Assessment, it improves macro F1 from 0.744 to 0.787. These findings reveal that explicit, source-grounded structure makes agent skills easier to search and review. They also suggest that SSL is best understood as a practical step toward more inspectable, reusable, and operationally actionable skill representations for agent systems, rather than as a finished standard or an end-to-end mechanism for managing and using skills.
Tsinghua AI