TY - JOUR TI - WikiPlanner: wiki-derived temporal monitors for safe LLM agent planning - a medical-domain proof of concept JO - AI Agent PY - 2026 VL - 2 IS - 3 SP - EP - 25 SN - ISSN 3070-3719 (Online) AB -

Large language model (LLM) agents in high-stakes sequential planning may produce locally reasonable steps whose overall sequence violates domain constraints; single-step filters cannot detect such cross-step temporal violations. We present WikiPlanner, which extracts temporal constraints from domain wikis, structures them via a typed Temporal Constraint Intermediate Representation (TC-IR) with seven constraint labels compiled to six Linear Temporal Logic over finite traces (LTLf) templates, and instantiates them as parallel runtime monitors. The two-stage pipeline - TC-IR extraction followed by deterministic template compilation - isolates LLM extraction uncertainty from formal compilation. We evaluate every stage on a medical planning benchmark of 45 constraints and 236 test plans (137 aligned, 99 natural). A dual-annotator audit of the 220 raw extractions (κ = 0.451, adjudicated) places semantic extraction precision at 25.9%; an automated check finds 79.5% of these records structurally well-formed (90.1% on a held-out nursing corpus); invented orderings, state-vs.-action confusion, and silent parse loss from output truncation are the dominant error classes. Plan validation achieves F1 = 0.920 (95%CI: 0.873-0.956) on the aligned set and F1 = 0.582 (95%CI: 0.447-0.701) on the natural set; the gap is attributed primarily to constraint-library coverage and symbol grounding, as wiki-derived constraint symbols and generator vocabulary form largely disjoint namespaces. Latency scales linearly with library size (0.006-0.059 ms). Single-pass template repair resolves 39.1% of violation plans (mean 1.30 word-level edits per success); local post-trigger repairs substantially outperform pre-trigger modifications. Results are a single-domain proof of concept, and symbol grounding is the main obstacle to natural-plan deployment.

KW - LLM agent planning KW - temporal logic KW - runtime verification KW - knowledge-based constraints KW - clinical decision support DO - 10.20517/aiagent.2026.22 UR - https://dx.doi.org/10.20517/aiagent.2026.22