{"ID":2834344,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2602.17046","arxiv_id":"2602.17046","title":"Dynamic System Instructions and Tool Exposure for Efficient Agentic LLMs","abstract":"Large Language Model (LLM) agents often run for many steps while re-ingesting long system instructions and large tool catalogs each turn. This increases cost, agent derailment probability, latency, and tool-selection errors. We propose Instruction-Tool Retrieval (ITR), a RAG variant that retrieves, per step, only the minimal system-prompt fragments and the smallest necessary subset of tools. ITR composes a dynamic runtime system prompt and exposes a narrowed toolset with confidence-gated fallbacks. Using a controlled benchmark with internally consistent numbers, ITR reduces per-step context tokens by 95%, improves correct tool routing by 32% relative, and cuts end-to-end episode cost by 70% versus a monolithic baseline. These savings enable agents to run 2-20x more loops within context limits. Savings compound with the number of agent steps, making ITR particularly valuable for long-running autonomous agents. We detail the method, evaluation protocol, ablations, and operational guidance for practical deployment.","short_abstract":"Large Language Model (LLM) agents often run for many steps while re-ingesting long system instructions and large tool catalogs each turn. This increases cost, agent derailment probability, latency, and tool-selection errors. We propose Instruction-Tool Retrieval (ITR), a RAG variant that retrieves, per step, only the m...","url_abs":"https://arxiv.org/abs/2602.17046","url_pdf":"https://arxiv.org/pdf/2602.17046v1","authors":"[\"Uria Franko\"]","published":"2025-12-01T06:43:43Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}