# Policy-as-Skill: Governed LLM Decision Support with Evidence, Deterministic Control, and Audit

arXiv: https://arxiv.org/abs/2609.27087 · HTML: https://mlsanity.com/paper/2609.27087/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 27 references that are in the map.
- First posted on arXiv 22 Sep 2026.

## Builds on (most cited first, 19 of 19)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning](https://mlsanity.com/paper/2209.14610/) (2022)
- [Prompt-Based Monte-Carlo Tree Search for Goal-Oriented Dialogue Policy Planning](https://mlsanity.com/paper/2305.13660/) (2023)
- [Ethical Reasoning and Moral Value Alignment of LLMs Depend on the Language we Prompt them in](https://mlsanity.com/paper/2404.18460/) (2024)
- [MORL-Prompt: An Empirical Analysis of Multi-Objective Reinforcement Learning for Discrete Prompt Optimization](https://mlsanity.com/paper/2402.11711/) (2024)
- [StablePrompt: Automatic Prompt Tuning using Reinforcement Learning for Large Language Models](https://mlsanity.com/paper/2410.07652/) (2024)
- [Policy-as-Prompt: Turning AI Governance Rules into Guardrails for AI Agents](https://mlsanity.com/paper/2509.23994/) (2025)
- [Ask more, know better: Reinforce-Learned Prompt Questions for Decision Making with Large Language Models](https://mlsanity.com/paper/2310.18127/) (2023)
- [Dialogue for Prompting: a Policy-Gradient-Based Discrete Prompt Generation for Few-shot Learning](https://mlsanity.com/paper/2308.07272/) (2023)
- [Privacy Policy Analysis through Prompt Engineering for LLMs](https://mlsanity.com/paper/2409.14879/) (2024)
- [Efficient Policy Adaptation with Contrastive Prompt Ensemble for Embodied Agents](https://mlsanity.com/paper/2412.11484/) (2024)
- [Using LLMs for Automated Privacy Policy Analysis: Prompt Engineering, Fine-Tuning and Explainability](https://mlsanity.com/paper/2503.16516/) (2025)
- [Do We Really Need SFT? Prompt-as-Policy over Knowledge Graphs for Cold-start Next POI Recommendation](https://mlsanity.com/paper/2510.08012/) (2025)
- [Hierarchical Prompt Decision Transformer: Improving Few-Shot Policy Generalization with Global and Adaptive Guidance](https://mlsanity.com/paper/2412.00979/) (2024)
- [Prompt Governance? On Governing Technologies Governed by Natural Language](https://mlsanity.com/paper/2606.07539/) (2026)
- [A Minimalist Prompt for Zero-Shot Policy Learning](https://mlsanity.com/paper/2405.06063/) (2024)
- [Prompt Commons: Collective Prompting as Governance for Urban AI](https://mlsanity.com/paper/2509.12415/) (2025)

## Abstract (arXiv)

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.