Skill benefits for LLM agents largely disappear in realistic retrieval settings from a 34k skill pool, approaching no-skill baselines, though query-specific refinement recovers much of the lost performance.
Each subdirectory contains a skill with a SKILL.md and possibly supporting files (scripts, references, etc.)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
How Well Do Agentic Skills Work in the Wild: Benchmarking LLM Skill Usage in Realistic Settings
Skill benefits for LLM agents largely disappear in realistic retrieval settings from a 34k skill pool, approaching no-skill baselines, though query-specific refinement recovers much of the lost performance.