arXiv:2609.27087v1 Announce Type: new Abstract: Organizations increasingly use LLMs for policy, compliance, risk, and operational decision support, requiring evidence validation, review routing, version control, and auditability. We introduce Policy-as-Skill (PaS), a modular runtime that packages these functions as executable, versioned policy capabilities. Thirteen methods are evaluated with a fixed Gemma4 backend on 600 development tasks. PaS+Audit achieves 53.8% exact accuracy, macro-F1 0.346, review F1 0.854, citation precision 1.000, policy-reference recall 0.984, and audit completeness 1.000, outperforming LLM+RAG on most governance and review metrics. Deterministic control raises aggregate accuracy to 61.2% but is strongly task dependent, supporting selective rather than universal rule-based intervention.
Policy-as-Skill: Governed LLM Decision Support with Evidence, Deterministic Control, and Audit
A new arXiv paper (2609.27087v1) introduces Policy-as-Skill (PaS), a modular runtime that packages evidence validation, review routing, version control, and auditability as executable, versioned policy capabilities for LLM-based policy, compliance, risk, and operational decision support. Evaluating thirteen methods with a fixed Gemma4 backend on 600 development tasks, the authors report PaS+Audit reaches 53.8% exact accuracy, 0.346 macro-F1, 0.854 review F1, 1.000 citation precision, 0.984 policy-reference recall, and 1.000 audit completeness, outperforming LLM+RAG on most governance and review metrics. Deterministic control raises aggregate accuracy to 61.2% but is strongly task dependent, which the authors say supports selective rather than universal rule-based intervention.
Run your AI side-project on zahid.host
EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.