[arXiv Research] Agentic Workflow Improvements — 2026-08-22

[github-actions[bot]]

Summary

25 papers screened, 18 relevant, 3 actionable opportunities identified.

---

Actionable Opportunities

Token Optimization — Optimal Skill Selection Under Budget Constraints

Paper: [Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees]((arxiv.org/redacted)
Authors: Yu Chen, Ruishuo Chen, Xun Wang
Published: 2026-08-20
Effort: medium

Rationale: The paper models skill selection as maximizing monotone submodular benefit minus context penalty under hard token constraints. gh-aw currently loads skills lazily but uses simple heuristics; this optimization framework could replace ad-hoc skill loading with provably efficient selection that balances skill utility against token cost.

---

Workflow Compilation — Structured Validation Harness with Admissible-Analysis Rules

Paper: [Bringing analytic rigor to agentic AI for science: The Brain Researcher platform for neuroimaging data analysis]((arxiv.org/redacted)
Authors: Zijiao Chen, Nicholas Lu, Xinhui Li
Published: 2026-08-20
Effort: medium

Rationale: Brain Researcher enforces rules for admissible analyses, required checks, and claim scope limits that dramatically improve grounding and accuracy. gh-aw could adopt a similar validation layer in the compiler that declares admissible tool sequences, mandatory validation steps, and output claim boundaries for each workflow type.

---

Prompt Engineering — MCP Skill Training Corpus Synthesis

Paper: [MidTool: Mid-training Data Synthesis for Agentic Tool Use]((arxiv.org/redacted)
Authors: Fengqing Jiang, Yite Wang, Boyi Liu
Published: 2026-08-20
Effort: high

Rationale: MidTool demonstrates synthesized training data from real-world tool APIs and MCP skills that teach models to recognize tool capabilities and compose workflows. This could inform how gh-aw structures SKILL.md files and agent instruction patterns to maximize model comprehension of available tools.

---

Papers Analyzed

Paper	Published	Relevant	Area
[An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction]((arxiv.org/redacted)	2026-08-20	No	—
[Inducing Task Models from Computer-Use Traces]((arxiv.org/redacted)	2026-08-20	Yes	—
[AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvement]((arxiv.org/redacted)	2026-08-20	Yes	—
[MidTool: Mid-training Data Synthesis for Agentic Tool Use]((arxiv.org/redacted)	2026-08-20	Yes	prompt-engineering
[Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents]((arxiv.org/redacted)	2026-08-20	Yes	—
[Growth Without Us: Machine Consumers, Corporate Circularity, and the Decoupling of GDP from Humanity after AGI]((arxiv.org/redacted)	2026-08-20	No	—
[The Third Restructuring of Software Form: From the Three-Tier Architecture to Storage, Models, and Agents]((arxiv.org/redacted)	2026-08-20	Yes	—
[From Agent Behaviour to Agent-Friendly Documentation: An Empirical Study of How Coding Agents Discover, Read, and Write Technical Documentation]((arxiv.org/redacted)	2026-08-20	Yes	—
[Task-CoEvolve: Efficient Harness Optimization via Adaptive Validation Task Selection]((arxiv.org/redacted)	2026-08-20	Yes	—
[Reward-Guided Autoregressive Graph Generation for Efficient Multi-Agent Communication Topology Design]((arxiv.org/redacted)	2026-08-20	Yes	—
[A three-dimensional typology of agency for advanced AI systems]((arxiv.org/redacted)	2026-08-20	No	—
[Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees]((arxiv.org/redacted)	2026-08-20	Yes	token-optimization
[ReguSim: Evaluating LLM Agent Rule Grounding in Financial Compliance]((arxiv.org/redacted)	2026-08-20	Yes	—
[G-MARK: Grounded Multi-Agent Reasoning for Cooperative Driving via Knowledge Graphs]((arxiv.org/redacted)	2026-08-20	No	—
[Bringing analytic rigor to agentic AI for science: The Brain Researcher platform for neuroimaging data analysis]((arxiv.org/redacted)	2026-08-20	Yes	workflow-compilation
[MaliciousSkillBench: A Comprehensive Benchmark for Malicious Agent Skill Detection]((arxiv.org/redacted)	2026-08-20	Yes	—
[TESTNAV: Pareto-Guided Search for Compositional Robustness Testing]((arxiv.org/redacted)	2026-08-20	No	—
[EnvHarness: Awakening Static Worlds for Agent Learning]((arxiv.org/redacted)	2026-08-20	Yes	—
[A knowledge-guided agentic framework for mitigating patient-context ambiguity in health queries]((arxiv.org/redacted)	2026-08-20	No	—
[PolicyGuide: From Guarding One Action to Guiding the Whole Workflow for Policy-Compliant LLM Agents]((arxiv.org/redacted)	2026-08-20	Yes	—
[Inadvertent Context Leakage in Language Models]((arxiv.org/redacted)	2026-08-20	Yes	—
[Repo0: Design-Driven Zero-to-All Code Generation]((arxiv.org/redacted)	2026-08-20	Yes	—
[SAPO: Single-Rollout Autoregressive Policy Optimization for Agentic Reinforcement Learning]((arxiv.org/redacted)	2026-08-20	Yes	—
[Adaptive Probabilistic Shielding by Learning MDPs for Safe Reinforcement Learning]((arxiv.org/redacted)	2026-08-20	No	—
[MileGPO: Milestone Inference with Local Evidence for Graph-Based Policy Optimization of Long-Horizon LLM Agents]((arxiv.org/redacted)	2026-08-20	Yes	—

---

Next Steps

• Investigate: Optimal skill selection algorithm (effort: medium)
• Investigate: Structured validation harness for workflows (effort: medium)
• Investigate: MCP skill corpus synthesis patterns (effort: high)

Generated by 🔬 arXiv Paper Researcher: GitHub Agentic Workflows · sonnet45 · 87.9 AIC · ⊞ 3.9K · ◷

• expires on Sep 5, 2026, 12:42 AM UTC-08:00