Agentic Harnesses: LLM-Driven Verification Layers for Robot Autonomy

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Rohan Bhagra, Mahantesh Halapannavar, Uddhav Bhattarai
• Venue: arXiv: cs.RO
• Published: 2026-08-10
• Collected: 2026-08-11
• URL: http://arxiv.org/abs/2608.09857v1
• PDF: https://arxiv.org/pdf/2608.09857v1
• DOI: UNKNOWN
• Registry group: academic_apis_and_preprints
• Evidence role: canonical_academic_index
• Language: en
• Country / region: UNKNOWN

Plain-Language Brief

Researchers propose a verification layer driven by large language models (LLMs) to evaluate the feasibility and ethics of actions suggested by robot planning models before they are executed. The system aims to reduce risks associated with biased goals, safety issues, and vulnerability to adversarial attacks.

What the Document Says

• The research focuses on a verification layer that utilizes LLMs to assess the permissibility of actions proposed by robot planning models.
• The proposed system, called LLM-as-a-Judge ensemble, uses chain-of-thought reasoning across multiple models and combines their outputs, mirroring a mixture of experts and self-consistency approach.
• This layer serves as middleware, filtering plans from the server's planning module before they reach the MCP server and the robot's low-level controls.
• The system achieves near 85% precision across accept/escalate/reject categories, contains 97% of adversarial attacks, and has negligible errors between accepting and rejecting tasks, with most errors occurring at the escalate boundary.

Key Concepts Mentioned

• robot autonomy
• verification layer
• large language models (LLMs)
• chain-of-thought reasoning
• adversarial attacks

Claims Made by the Document

• The proposed system can evaluate the permissibility of actions.
• The system achieves near 85% precision across accept/escalate/reject categories.
• The system contains 97% of adversarial attacks.

Evidence Used

• ...a LLM-driven verification layer between planning and execution to evaluate action permissibility...

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Advances in advanced artificial intelligence tools have sparked research in robot autonomy, but the development of such systems has largely focused on execution rather than verifying the feasibility actions planning models propose. Like general-purpose LLMs, robotics planning models carry risks: biased toward user-specified goals, they may suggest actions misaligned with scientific ethics, they may be unsafe due to an inability to 'remember' prior safety risks, or they may be vulnerable to adversarial attacks on the autonomy ecosystem. We propose a LLM-driven verification layer between planning and execution to evaluate action permissibility. Our LLM-as-a-Judge ensemble combines chain-of-thought reasoning across models and synthesizes those expert judge outputs, mirroring a combination of a mixture of experts and self-consistency approach. This layer serves as middleware, gating plans from the server's planning module before they reach the MCP server and therefore the robot's low-level controls: plans are approved, rejected for reformulation, or escalated for human review. With this system, we achieve near 85% precision across accept/escalate/reject categories 97% containment of adversarial attacks, with negligible errors between accepting and rejecting tasks, and errors mostly manifesting at the escalate boundary.