Harnessing LLM Agents with Skill Programs

TL;DR

HASP introduces executable skill programs for LLM agents, enabling explicit intervention and correction, leading to significant performance improvements across web-search, math reasoning, and coding tasks.

Contribution

The paper presents HASP, a modular framework that upgrades skills into executable functions for improved intervention, supervision, and self-improvement in LLM agents.

Findings

Inference-time PFs improve web-search reasoning by 25%.

Post-training and evolution methods achieve 30.4% gain over baseline.

HASP enhances performance on math reasoning and coding tasks.

Abstract

Equipping LLM agents with reusable skills derived from past experience has become a popular and successful approach for tackling complex and long-horizon tasks. However, such lessons are often encoded as textual guidance that remains largely advisory, lacking explicit mechanisms for when and how to intervene in the agent loop. To bridge the gap, we introduce HASP(Harnessing LLM Agents with Skill Programs), a new framework that upgrades skills into executable Program Functions (PFs). Rather than offering passive advice, PFs act as executable guardrails that activate on failure-prone states and modify the next action or inject corrective context. HASP is highly modular: it can be applied at inference time for direct agent-loop intervention, during post-training to provide structured supervision, or for self-improvement by evolving validated, teacher-reviewed PFs. Empirically, HASP drives substantial gains compared to both training-free and training-based methods on web-search, math reasoning, and coding tasks. For example, on web-search reasoning, inference-time PFs alone improve the average performance by 25% compared to (multi-loop) ReAct Agent, while post-training and controlled evolution achieve a 30.4% gain over Search-R1. To provide deeper insights into HASP, our mechanism analysis reveals how PFs trigger and intervene, how skills are internalized, and the requirement for stable skill library evolution.