Generative Skill Composition
Free while signed in. Answers cite the passages they came from.
Coding agents accumulate large skill libraries, and picking the right skills for a task has become the bottleneck. The usual options either dump the whole collection into context or retrieve skills with embeddings and rerankers, and both treat selection as a ranking problem rather than a joint plan. ---
Composition as one joint decision: SkillComposer decides which skills, how many, and in what order all at once, instead of scoring skills independently and hoping the pieces fit together.
A constrained autoregressive decoder: A decoder over skill identifiers produces the full plan in a single pass, so dependencies between successive skills fall out of the generation naturally.
Strong gains at lower token cost: On SkillsBench with frontier models, it lifts pass rate well beyond the no-skill baseline, beats top-3 retrieval, and matches the gold-skill upper bound while using fewer prompt tokens.
Why it matters: As skill libraries keep growing, treating selection as generation rather than retrieval is what lets agents surface and sequence the right capabilities without drowning in their own toolbox.
Get next week’s papers.
The same picks and the same summaries, in your inbox. Free, and 176 issues deep.
Subscribe on Substack