Shepherd: A Runtime Substrate Empowering Meta-Agents with a Formalized Execution Trace

• URL: http://arxiv.org/abs/2605.10913v1
• Date: Mon, 11 May 2026 17:50:51 GMT
• Title: Shepherd: A Runtime Substrate Empowering Meta-Agents with a Formalized Execution Trace
• Abstract summary: We introduce Shepherd, a functional programming model that formalizes meta-agent operations on target agents as functions.<n>Shepherd records every agent-environment interaction as a typed event in a Git-like execution trace, enabling any past state to be forked and replayed.<n>The system forks the agent process and its $5$ faster than Docker, achieving $>95%$ prompt-cache reuse on replay.
• Score: 37.66254228361724
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: We introduce Shepherd, a functional programming model that formalizes meta-agent operations on target agents as functions, with core operations mechanized in Lean. Shepherd records every agent-environment interaction as a typed event in a Git-like execution trace, enabling any past state to be forked and replayed. The system forks the agent process and its filesystem $5\times$ faster than Docker, achieving $>95\%$ prompt-cache reuse on replay. We demonstrate the model through three applications. First, in runtime intervention, a live supervisor increases pair coding pass rates from 28.8% to 54.7% on CooperBench. Second, in counterfactual meta-optimization, branching exploration outperforms baselines across four benchmarks by up to 11 points while reducing wall-clock time by up to 58%. Third, in Tree-RL training, forking rollouts at selected turns improves TerminalBench-2 performance from 34.2% to 39.4%. These results establish Shepherd as an efficient infrastructure for programming meta-agents. We open-source the system to support future research.

Related papers

• Grounding Agent Memory: Environment-Probing Curation for Enterprise Agents [37.62160903348546]
  We introduce environment-probing curation, a deployment-compatible extension that gives an existing asynchronous curator agent least-privilege.<n>It requires no model retraining and leaves the task agent, retriever, memory representation, and production write authority unchanged.<n>On CLBench, probing raises pass rate from 39% to 73% and pass-discounted reward from 8.60 to 22.60 while reducing queries from 8.8 to 4.7 per question.
  arXiv  Detail & Related papers  (2026-09-10T04:06:16Z)
• Closing the Consistency Gap: Self-Evolving Agents That Learn to Stay on Course [5.9756241643216255]
  Large language model (LLM)-powered agents can be accurate on average yet unreliable in production.<n>We present a self-evolving framework that reduces this gap by identifying unstable, low-consistency steps in agent trajectories.<n>On AppWorld with ReAct/GPT-4.1, our framework raises the fraction of tasks that succeed in all five runs by +16 points on same-task evaluation and +13 points on similar-task generalization.
  arXiv  Detail & Related papers  (2026-09-08T14:53:43Z)
• Recursive Synthesis for Long-Horizon Terminal Tasks [50.39352383646813]
  Recursive Synthetic Terminal Tasks (RST) is a framework for constructing long-horizon terminal-agent tasks at scale.<n>RST produces 37,484 synthesized terminal-agent tasks at roughly $0.05 per task.<n>Fine-tuning on trajectories improves Qwen3.5-27B and Qwen3.5-122B-A10B by up to 10 points on Terminal-Bench 2, Terminal-Bench Hard, and Long-Horizon Terminal Bench.
  arXiv  Detail & Related papers  (2026-08-05T23:24:26Z)
• VisualClaw: A Real-Time, Personalized Agent for the Physical World [76.89211120690028]
  We present VisualClaw, a self-evolving multimodal agent built around two principles.<n>First, hybrid encoding reduces deployment cost by filtering less informative streaming frames with a cascaded gate.<n>Second, skill evolution lets the agent learn from failures, producing skill-bank updates that help future questions.
  arXiv  Detail & Related papers  (2026-06-15T06:58:22Z)
• ISE: An Execution-Grounded Recipe for Multi-Turn OS-Agent Trajectories [18.739245125968175]
  Training capable OS agents requires data that simultaneously captures structured user intents, multi-turn task delegation, and grounded tool execution--properties absent from existing datasets.<n>We propose ISE (Intent -> Simulate -> Execute), a three-stage synthesis paradigm that addresses these gaps jointly.
  arXiv  Detail & Related papers  (2026-06-09T23:44:26Z)
• A Policy-Driven Runtime Layer for Agentic LLM Serving [6.975319765760388]
  Multi-agent LLM systems have become the dominant production workload, but the serving stack was not built for them.<n>We argue this seam is best addressed by an architectural change rather than point fixes.<n>Preliminary results on five real multi-agent workloads show +13 to +37 pp cache hit-rate lift, 12% to 29% lower mean TTFT, and 6% to 14% higher throughput over an unmodified serving stack.
  arXiv  Detail & Related papers  (2026-05-26T22:38:34Z)
• WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation [88.10947115397971]
  This work presents WildClawBench, a native-runtime benchmark of 60 human-authored, bilingual, multimodal tasks spanning six thematic categories.<n>Each task averages roughly 8 minutes of wall-clock time and over 20 tool calls.<n> Grading is hybrid, combining deterministic rule-based checks, environment-state auditing of side effects, and an LLM/VLM judge for semantic verification.
  arXiv  Detail & Related papers  (2026-05-11T17:49:43Z)
• Workspace Optimization: How to Train Your Agent [49.579258311548635]
  We argue it is the agent's emphworkspace, the structured external substrate it reads, writes, and tests.<n>We propose a principled way to evolve the workspace, mirroring the structure of weight-space training.<n>We instantiate the idea in DreamTeam, a multi-agent harness for ARC-AGI-3.
  arXiv  Detail & Related papers  (2026-05-10T16:52:10Z)
• ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents [77.22389710754452]
  We introduce bench, a benchmark for coworker agents built around multi-turn multi-day tasks.<n>The current release contains 100 tasks across 13 professional scenarios, executed against five stateful sandboxed services.<n>The strongest model reaches 75.8 weighted score, but the best strict Task Success is only 20.0%, indicating that partial progress is common.
  arXiv  Detail & Related papers  (2026-04-26T16:05:02Z)
• Deep Researcher Agent: An Autonomous Framework for 24/7 Deep Learning Experimentation with Zero-Cost Monitoring [2.4112990554464235]
  We present textbfDeep Researcher Agent, an open-source framework that enables large language model (LLM) agents to autonomously conduct deep learning experiments around the clock.<n>Unlike existing AI research assistants that focus on paper writing or code generation, our system addresses the full experiment lifecycle: hypothesis formation, code implementation, training execution, result analysis, and iterative refinement.
  arXiv  Detail & Related papers  (2026-04-07T13:16:31Z)
• ClawsBench: Evaluating Capability and Safety of LLM Productivity Agents in Simulated Workspaces [17.202580606345666]
  Large language model (LLM) agents are increasingly deployed to automate productivity tasks.<n>Existing benchmarks rely on simplified environments and fail to capture realistic, multi-service environments.<n>We introduce ClawsBench, a benchmark for evaluating and improving LLM agents in realistic productivity settings.
  arXiv  Detail & Related papers  (2026-04-06T21:09:06Z)
• WebGen-Agent: Enhancing Interactive Website Generation with Multi-Level Feedback and Step-Level Reinforcement Learning [51.14454312533818]
  WebGen-Agent is a novel website-generation agent that leverages comprehensive and multi-level visual feedback.<n>We introduce textitStep-GRPO with Screenshot and GUI-agent Feedback to improve the ability of LLMs to act as the reasoning engine of WebGen-Agent.<n>WebGen-Agent increases the accuracy of Claude-3.5-Sonnet from 26.4% to 51.9% and its appearance score from 3.0 to 3.9, outperforming the previous state-of-the-art agent system.
  arXiv  Detail & Related papers  (2025-09-26T17:59:51Z)
• GitGoodBench: A Novel Benchmark For Evaluating Agentic Performance On Git [0.8397730500554048]
  GitGoodBench is a novel benchmark for evaluating AI agent performance on Version Control System (VCS) tasks.<n>Our benchmark covers three core Git scenarios extracted from open-source Python, Java, and Kotlin repositories.<n>We establish baseline performance on the prototyping version of our benchmark using GPT-4o equipped with custom tools, achieving a 21.11% solve rate overall.
  arXiv  Detail & Related papers  (2025-05-28T16:56:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.