aiqbench analysis
The paper argues that LLM agents should query a typed skill graph through an interface, rather than receive a precomputed retrieval bundle. Its reported gains, including +14.3 points on ALFWorld, come from one benchmark family and three models, so they need independent replication before generalizing. For practitioners building large tool or skill libraries, the more durable point is that inter-skill relations such as prerequisites, conflicts, and duplicates matter for selection, and that letting agents revise those relations may matter as much as the retriever.
SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale
arXiv:2606.03056v3 Announce Type: replace Abstract: Large skill libraries are now standard equipment for LLM agents, and which subset reaches the prompt decides what the agent can do. Enumeration and top-$K$ embedding retrieval both treat skills as independent documents, assuming relevance is a property of one skill at a time. It is not: skills depend on, conflict with, specialize, and duplicate one another, so the closest match to a task can name it while omitting a prerequisite it needs. Graph retrieval encodes these relations but spends them in a fixed pipeline, handing the agent an opaque bundle. We present SkillDAG, which exposes a typed skill graph to the LLM as an agent-callable interface rather than consuming it for the agent: search returns vector matches, typed-edge neighbors, and conflict signals, and a propose-then-commit protocol lets it register execution-backed edges as it works. Across two benchmarks and three model families, SkillDAG outperforms full-library prompting, vector retrieval, and Graph-of-Skills, gaining +14.3 points of ALFWorld success and tying only at ceiling; its ranking holds flat across a $10\times$ pool expansion that degrades the fixed pipeline. Inter-skill structure, these results suggest, is worth more to an agent that can query and revise it than to a retriever that consumes it.
Abstract:Large skill libraries are now standard equipment for LLM agents, and which subset reaches the prompt decides what the agent can do. Enumeration and top-$K$ embedding retrieval both treat skills as independent documents, assuming relevance is a property of one skill at a time. It is not: skills depend on, conflict with, specialize, and duplicate one another, so the closest match to a task can name it while omitting a prerequisite it needs. Graph retrieval encodes these relations but spends them in a fixed pipeline, handing the agent an opaque bundle. We present SkillDAG, which exposes a typed skill graph to the LLM as an agent-callable interface rather than consuming it for the agent: search returns vector matches, typed-edge neighbors, and conflict signals, and a propose-then-commit protocol lets it register execution-backed edges as it works. Across two benchmarks and three model families, SkillDAG outperforms full-library prompting, vector retrieval, and Graph-of-Skills, gaining +14.3 points of ALFWorld success and tying only at ceiling; its ranking holds flat across a $10\times$ pool expansion that degrades the fixed pipeline. Inter-skill structure, these results suggest, is worth more to an agent that can query and revise it than to a retriever that consumes it.
Artificial Intelligence (cs.AI)
arXiv:2606.03056 [cs.AI]
(or arXiv:2606.03056v3 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2606.03056 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 291 words
Related coverage
SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale
arXiv · Jun 3, 2026
SE-GoS: Self-Evolving Graph-of-Skills for Skill Library at Scale
arXiv · Sep 9, 2026
EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-Evolution
arXiv · Oct 8, 2026
AXRP Episode 50 - Eli Lifland on AI 2027
LessWrong · Aug 3, 2026