cs.LG提交于 2026-10-19已译

多步智能体工具推荐：找到任务所需的每一个工具

Tool Recommendation for Multi-Step Agents: Finding Every Tool a Task Needs

Sam Partee · River Stevens · Thierry Damiba · Valerie Fanelle

中文摘要

多步智能体任务常因缺失单个工具而失败，但所需工具数量庞大，约 8,200 个工具的定义超出模型上下文窗口。Tool Recommendation 系统通过三个固定接口工具，在智能体与工具网关间运行，使用户凭证执行调用并检查服务同意。在 ToolRet 基准上，其 Top-10 召回率比 BM25 高 57%，性能优于规模大九倍的公开系统。在配对多步工作流中，93% 的 Codex 会话首次推荐即覆盖所有必需工具。相比固定 100 工具列表，Codex 输入 token 减少 17%，Claude Code 减少 40%，14 次运行总账单最高降低 40%，且任务完成率无变化。

关键要点

01解决多步智能体因工具缺失导致任务失败，且工具定义超出模型上下文窗口的难题。
02部署 Tool Recommendation 系统，通过三个固定接口工具在网关层执行调用并验证用户权限。
03在 ToolRet 基准上，Top-10 召回率比 BM25 高 57%，性能优于规模大九倍的公开系统。
0493% 的 Codex 会话首次推荐即覆盖所有必需工具，显著降低输入 token 和总账单成本。

解读

尚无解读。

原始英文摘要

arXiv:2610.23707v1 Announce Type: new Abstract: A multi-step agent task fails when a single tool it needs is missing, yet the tools an agent may need are too many to list: the roughly 8,200 tools on the platform we study have definitions that together exceed the context window of any model we evaluate. We present Tool Recommendation, a deployed system that gives an agent all of them, hundreds of times more than a typical MCP server exposes, through a fixed interface of three tools. It runs at the gateway between agent and tools, executing every call under the end user's own credentials and checking it against the user's consent for each service and the tools the agent may use. We evaluate it on both a public retrieval benchmark and billed end-to-end session cost. On ToolRet, Tool Recommendation places every tool a query needs in the top ten 57% more often than BM25, a lexical ranking that commercial tool search offers, and compares favorably with published systems up to nine times its size. In paired multi-step workflows, its first recommendation covered every required tool in 93% of Codex sessions, and against a fixed list of 100 tools it cut input tokens by 17% with Codex and 40% with Claude Code and the total bill in all 14 runs, by up to 40%, with no change in task completion.