Gavel reads skill-routing signals from a frozen model

Gavel reads task and skill representations from a frozen Qwen3-32B with two trained linear maps. It shortlists skills without loading the whole catalog into the prompt, then resumes model inference to judge candidates. On public benchmarks and a new 372-trajectory set, it improves routing by up to 13.4 points on written tasks and 21.9 points during agent rollouts against tested disclosure and retrieval baselines. No public implementation is linked yet.

For teams with large skill catalogs, the result suggests routing can reuse a model’s internal signal and save prompt space. Trial only after code and held-out library tests are available; the reported comparison is still author-run.