SkillCorpus: Consolidating and Evaluating the Open Skill Ecosystem for Real-World LLM Agents

Researchers introduce SkillCorpus, a framework that aggregates, curates, matches, and evaluates open-source SKILL.md files for LLM agents, filtering ~821,000 crawled skills into 96,401 organized by a 16-class taxonomy and three quality facets. Integrating SkillCorpus yields consistent gains across three benchmarks, with the largest improvement of +7.5 percentage points on SkillsBench. The work provides the first end-to-end account of when a curated, retrieval-served community corpus improves real agent tasks.

arXiv:2607.15557v1 Announce Type: new Abstract: Agent skills, SKILL.md files that package reusable procedural knowledge for an LLM agent, are a popular mechanism for extending agent capabilities. Public repositories now host them in large and growing numbers, yet these artifacts are fragmented, redundant, and uneven in quality, and their value in practice is unclear. A core question remains open, namely how to consolidate this open-source SKILL.md ecosystem into a single usable corpus, and what bounds its benefit on real-world agent tasks. We present SkillCorpus, a framework that aggregates, curates, matches, and evaluates the open skill ecosystem at scale. It filters ~821,000 crawled skills through a multi-stage pipeline into 96,401 skills organised by a 16-class taxonomy and three quality facets utility, robustness, safety , and pairs them with a fine-tuned retrieval-and-selection stack that matches task-relevant skills. We evaluate end-to-end across three benchmarks SkillsBench, GDPVal, QwenClawBench , two harnesses, and two open backbones with a frontier robustness check. Integrating SkillCorpus yields consistent gains across all three benchmarks, largest on SkillsBench +7.5 pp . An operational analysis traces the gains to a coverage boundary and a harness boundary. SkillCorpus is, to our knowledge, the first end-to-end account of when a curated, retrieval-served community corpus improves real agent tasks, and where it does not. The dataset, models, and code will be released upon acceptance.