Know When to Trust the Skill: Delayed Appraisal and Epistemic Vigilance for Single-Agent LLMs

Computer Science > Artificial Intelligence [Submitted on 17 Apr 2026] Know When to Trust the Skill: Delayed Appraisal and Epistemic Vigilance for Single-Agent LLMs Eren Unlu As large language models (LLMs) transition into autonomous agents integrated with extensive tool ecosystems, traditional routing heuristics increasingly succumb to context pollution and "overthinking". We argue that the bottleneck is not a deficit in algorithmic capability or skill diversity, but the absence of disciplined second-order metacognitive governance. In this paper, our scientific contribution focuses on the computational translation of human cognitive control - specifically, delayed appraisal, epistemic vigilance, and region-of-proximal offloading - into a single-agent architecture. We introduce MESA-S (Metacognitive Skills for Agents, Single-agent), a preliminary framework that shifts scalar confidence estimation into a vector separating self-confidence (parametric certainty) from source-confidence (trust in retrieved external procedures). By formalizing a delayed procedural probe mechanism and introducing Metacognitive Skill Cards, MESA-S decouples the awareness of a skill's utility from its token-intensive execution. Evaluated under an In-Context Static Benchmark Evaluation natively executed via Gemini 3.1 Pro, our early results suggest that explicitly programming trust provenance and delayed escalation mitigates supply-chain vulnerabilities, prunes unnecessary reasoning loops, and prevents offloading-induced confidence inflation. This architecture offers a scientifically cautious, behaviorally anchored step toward reliable, epistemically vigilant single-agent orchestration. Comments: 7 pages, 1 figure Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2604.16753 [cs.AI]   (or arXiv:2604.16753v1 [cs.AI] for this version)   https://doi.org/10.48550/arXiv.2604.16753 Focus to learn more Submission history From: Eren Unlu Ph. D. [view email] [v1] Fri, 17 Apr 2026 23:55:19 UTC (129 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev   |   next > new | recent | 2026-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)