SkillSelect-Serve: Budget-Controllable and QoS-Aware Skill Service Recommendation and Composition for Small LLM Agents

2026-05-08

AI Summary

This work addresses the challenge faced by small-scale large language model agents in efficiently selecting reusable skill services under constraints of limited context length, tool availability, and risk tolerance. It introduces a novel formulation that models skills as structured service units with explicit awareness of quality-of-service (QoS) attributes and deployment constraints. The authors propose a recommendation framework operating under a strict service budget, which ranks candidate skills using a task-condition compatibility evaluator and enforces feasibility through a constraint projection mechanism that accounts for context length, cumulative risk exposure, and tool compliance. Evaluated on a repository of 35,353 skills, the approach achieves 100% task deliverability with only a 1.14-point hit rate penalty, reduces risk exposure by 50%, eliminates 44–81% of tool violations, and improves hit rate from 0.8864 to 0.9091 under a three-service budget.

Abstract

Reusable skill libraries are becoming important infrastructure for large language model (LLM) agents, yet existing selection methods often treat skills as retrievable documents and return fixed top-k lists. This paper presents SkillSelect-Serve, a budget-controllable and QoS-aware framework that formulates agent skill selection as Skill Service Recommendation and Composition. SkillSelect-Serve represents raw skills as structured Skill Services with functional descriptions, dependencies, context cost, risk, and QoS-related attributes. A local Micro-Agent Requirement Planner converts natural-language tasks into structured service requirements, while a shared discovery backbone retrieves candidate services from a large registry. The framework then performs dual-granularity utility modeling with skill-level marginal suitability estimation and bundle-level calibration for coverage, redundancy, cost, and risk trade-offs. Experiments on 35,353 skills and 586 task queries show that SkillSelect-Serve consistently improves same-budget bundle recall and mean utility over fixed top-k retrieval baselines.

Problem

• Skill Service Recommendation
• QoS-aware
• Budget-constrained
• LLM Agents
• Operational Constraints

Innovation

• QoS-aware recommendation
• budget-constrained selection
• structured skill services
• constrained projection
• LLM agent skills