Skip to content
Preprint

Compile, Then Page: Executable SOP Programs and a Capability-Gated Runtime for Procedural LLM Agents

Jul 2026 · 1 citation · 17 references
Computer Science

TL;DR

Enterprise agents must follow long-horizon, conditional, safety-critical standard operating procedures (SOPs), which are compiled into executable pseudo-code and run with a program-guided stack machine that pages the active frame while an LLM performs semantic execution.

Abstract

Enterprise agents must follow long-horizon, conditional, safety-critical standard operating procedures (SOPs). We compile machine-readable SOP constraints into executable pseudo-code and run them with a program-guided (PG) stack machine that pages the active frame while an LLM performs semantic execution. A three-arm SOPBench study across six models separates representation from runtime: compiled text never significantly hurts and gains up to 16.0 points where official prose underperforms. Runtime guidance is capability-gated. Two strong models independently show positive seven-domain PG contrasts (58:19 and 75:31 discordant pairs), whereas weak models are harmed. A full-program cursor ablation (active frame first, complete program retained) recovers much of the strong-model refusal gain; selective visibility adds a smaller improvement. Paired probe and audit measurements track this divide to spontaneous state discipline rather than reconstruction ability. On Bank the three primary arms rise from 70.4 to 86.4 to 92.8, with 100% refusal correctness. Practical guidance: compile first; enable active-frame paging only after a model-level discipline check.

View source

Similar papers

Preprint Jul 2026

Auto: The AGI Compiler

Auto is presented, a compiler that records live agent behavior, measures which parts are secretly deterministic, extracts them into verified programs or distilled specialists, and emits cognition binaries: WebAssembly artifacts whose manifests carry measured guarantees and whose declared capabilities are physically enforced by the sandbox.

Jaber Jaber, Osama Jaber · 1 citation
Preprint Jul 2026

IFHierBench: Hierarchical Instruction Following for Large Language Models

IFHierBench is introduced, a hierarchical instruction-following benchmark of 600 prompts stratified across four constraint-tree depths and 35 distinct constraints, each prompt paired with a deterministic checker that verifies satisfaction at every scope.

Yuetian Mao, Chunyang Chen · 0 citations
Preprint Jul 2026

CARE: Pre-Execution Command Verification for Shell-Executing LLM Agents

Large Language Model (LLM) agents are increasingly used for coding and terminal automation, making shell-command dispatch a high-stakes runtime control point. We study command-level pre-execution mediation for individual shell commands produced by LLM agents under bounded path context. Existing safeguards remain limited: generic guardrails do not model shell structure in sufficient detail, always-on LLM judges are relatively costly and variable, and shell parsers do not directly prevent harmful execution. We present CARE (Canonicalization, Attribution, and Resolution Engine), a shell-specific, static-first verifier for individual shell commands before execution. CARE canonicalizes generated commands into stable verification targets, derives deterministic evidence over syntax, command semantics, path context, and provenance-backed risk patterns, and escalates only underdetermined cases to an LLM judge. This design keeps the common case fast, reproducible, and auditable while reserving neural adjudication for borderline commands. On the balanced main split, CARE reaches 85.64% F1 with a 0.91% false-positive rate at 2.32 ms mean latency. When deployed in its static enforcement profile, CARE retains 84.99% F1 at 0.34 ms and reduces realised harm on RedCode-gen to 37.33%. Across external-generalization tests and controlled Docker-sandbox execution, these profiles expose a practical trade-off between benign recovery, false-positive burden, latency, and harm reduction. Overall, command-level shell mediation can reduce dispatch-boundary risk for LLM agents while preserving most benign workflows.

Yu Liu, Wenxiao Zhang, Zhiwei Yang et al. · 1 citation
Preprint Jul 2026

SIGIL: Compiling Agent Skills into Typed Harnesses

Skill Compilation is introduced, realized in SIGIL, which compiles a prose skill into an executable harness, and is model-independent: the harness holds at 86% across two model generations while prose swings from 56% to 68%.

Jayanaka L. Dantanarayana, Savini Kashmira, Lingjia Tang et al. · 0 citations
Book Open access Aug 2026

IntentP4: Bridging P4 Temporal Specifications and Executable Network Tests

IntentP4 is presented, a formal-methods-aided pipeline that translates an operator's natural-language intent into a P4LTL specification and then into a replayable multi-packet test case, grounded throughout in compiler artifacts via a tool-queryable ProgramContext and gated by deterministic per-stage validators.

Ruonan Feng, Mingming Zhang, Yu Jiang et al. · 0 citations
Preprint Jul 2026

PerfAgent: Profiler-Guided Iterative Refinement for Repository-Level Code Optimization

PerfAgent is presented, a profiler-guided, verifier-in-the-loop workflow that gives an off-the-shelf coding agent the feedback needed to find real hotspots, improve beyond the first passing patch, and use profiler evidence rather than timing alone to decide what to optimize next.

Ryan Deng, Yuanzhe Liu, Bastian Lipka et al. · 2 citations