MIND-Skill: Quality-Guaranteed Skill Generation via Multi-Agent Induction and Deduction

TL;DR

MIND-Skill is a framework that automatically induces high-quality, generalizable skills from successful problem-solving trajectories using multi-agent induction and deduction, with robust quality guarantees.

Contribution

It introduces a novel multi-agent framework that automatically distills reusable skills from trajectories, ensuring quality through multiple loss functions and evaluation on unseen tasks.

Findings

MIND-Skill outperforms existing skill generation methods on AppWorld and BFCL-v3 datasets.

The framework guarantees skill quality via reconstruction, outcome, and rubric losses.

Skills induced are effective on unseen tasks, demonstrating generalizability.

Abstract

Large language model (LLM) powered AI agents have emerged as a promising paradigm for autonomous problem-solving, yet they continue to struggle with complex, multi-step real-world tasks that demand domain-specific procedural knowledge. Reusable agent skills, which encapsulate successful problem-solving strategies, offer a natural remedy by enabling agents to build on prior experience. However, curating such skills has largely remained a manual endeavor, requiring human experts to distill rich domain knowledge into actionable guidelines. In this work, we present $\textbf{M}$ulti-agent $\textbf{IN}$duction and $\textbf{D}$eduction for $\textbf{Skill}$s ($\textbf{MIND-Skill}$), a framework that automatically induces generalizable skills from successful trajectories with robust quality guarantees. MIND-Skill consists of an induction agent which is tasked to abstract reusable skills from successful trajectories, and a deduction agent which aims to reconstruct trajectories by following the induced skills. To guarantee the quality of the generated skills, we introduce a reconstruction loss that compares input and reconstructed trajectories, an outcome loss that enforces the correctness of the reconstructed trajectories, and a rubric loss that assesses the documentation quality and regularizes the abstraction level of the generated skills according to predefined criteria. These textual losses are jointly optimized with TextGrad, and the resulting skills are evaluated on held-out tasks unseen during optimization. Experiments on AppWorld and BFCL-v3 show that MIND-Skill consistently outperforms concurrent skill generation methods.