# hierarchical agent systems

A hierarchical agent system is a tiered multi-agent AI system in which higher-level agents delegate subtasks to lower-level agents that communicate to accomplish complex tasks.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Attacks and Defenses in Decentralized Multi-Agent Systems with Hop-by-Hop Task Decomposition and Delegation

DevFeed: [Attacks and Defenses in Decentralized Multi-Agent Systems with Hop-by-Hop Task Decomposition and Delegation](<https://devfeed.tech/articles/attacks-and-defenses-in-decentralized-multi-agent-systems-with-hop-by-hop-task-decomposition-and-delegation-146019.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.25056>)

Author: Matthew D. Laws, Maria Lazou, Alina Oprea, Cristina Nita-Rotaru

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [ai security](<https://devfeed.tech/topics/ai-security.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Cybersecurity](<https://devfeed.tech/topics/cybersecurity.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [attacks](<https://devfeed.tech/tags/attacks.md>), [availability](<https://devfeed.tech/tags/availability.md>), [cryptographic](<https://devfeed.tech/tags/cryptographic.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [decentralized](<https://devfeed.tech/tags/decentralized.md>), [delegation](<https://devfeed.tech/tags/delegation.md>), [execution](<https://devfeed.tech/tags/execution.md>), [integrity](<https://devfeed.tech/tags/integrity.md>), [latency](<https://devfeed.tech/tags/latency.md>), [llm](<https://devfeed.tech/tags/llm.md>), [malicious](<https://devfeed.tech/tags/malicious.md>), [memory](<https://devfeed.tech/tags/memory.md>), [mitigation](<https://devfeed.tech/tags/mitigation.md>), [model](<https://devfeed.tech/tags/model.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [multi-agent-systems](<https://devfeed.tech/tags/multi-agent-systems.md>), [overhead](<https://devfeed.tech/tags/overhead.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

The abstract reports that one malicious agent in a decentralized multi-agent LLM system can cause up to 100% task failure, raise token use by up to 7.7x and runtime by up to 3.2x, or exfiltrate up to 94% of relevant prior tasks. The proposed cryptographic protocol creates a verifiable execution trace, which an auditor analyzes to detect and attribute malicious behavior. On HiddenBench, evaluations with gpt-6.1-sol and Qwen3.8-27B-FP8 found negligible latency overhead and 98% detection and correct attribution of attacks, including attacks that complied with the protocol.

### Source excerpt

arXiv:2611.25056v1 Announce Type: new Abstract: Decentralized multi-agent LLM systems enable independently administered agents, each with its own operator, model, memory, and configuration, to collaborate on complex tasks without centralized coordination. Agents autonomously decompose tasks, execute subtasks and delegate remaining work to peers, enabling scalability and eliminating single points of failure. However, decentralization also introduces significant security challenges, as no single entity has complete visibility into task execution or the authority to enforce correct behavior across agents. In this paper, we systematize threats from malicious agents in decentralized multi-agent systems and propose mitigation techniques. We first design and evaluate attacks against integrity, availability, and confidentiality, demonstrating that a single malicious agent can cause up to 100% task failure, inflate token consumption and runtime by up to 7.7x and 3.2x, respectively, and exfiltrate up to 94% of relevant prior tasks from benign agents. We then present a cryptographic protocol that produces an irrefutable verified execution trace of every action taken in the system and develop an auditor that analyzes these traces to detect and attribute malicious behavior. Evaluations on HiddenBench using gpt-6.1-sol and Qwen3.8-27B-FP8 show that our protocol incurs negligible latency overhead, while the auditor detects and correctly attributes 98% of attacks, including those that comply with the protocol.

## ResearchFlow-Eval: Tracing the Execution Gap in End-to-End Multi-Claim Autoresearch

DevFeed: [ResearchFlow-Eval: Tracing the Execution Gap in End-to-End Multi-Claim Autoresearch](<https://devfeed.tech/articles/researchflow-eval-tracing-the-execution-gap-in-end-to-end-multi-claim-autoresearch-144962.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.18088>)

Author: Peng Jiang, Dongyang Li, Quanying Liu, Xiaoxuan Jia

Published: 2026-11-20T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Orchestration](<https://devfeed.tech/topics/orchestration.md>), [DeepFact-Bench](<https://devfeed.tech/topics/deepfact-bench.md>), [iteration](<https://devfeed.tech/topics/iteration.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-harnesses](<https://devfeed.tech/tags/agent-harnesses.md>), [anchor](<https://devfeed.tech/tags/anchor.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [autoresearch](<https://devfeed.tech/tags/autoresearch.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [completion](<https://devfeed.tech/tags/completion.md>), [continuous](<https://devfeed.tech/tags/continuous.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [design](<https://devfeed.tech/tags/design.md>), [dimensions](<https://devfeed.tech/tags/dimensions.md>), [eval](<https://devfeed.tech/tags/eval.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [evaluations](<https://devfeed.tech/tags/evaluations.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [framework](<https://devfeed.tech/tags/framework.md>), [gap](<https://devfeed.tech/tags/gap.md>), [improvements](<https://devfeed.tech/tags/improvements.md>), [tracing](<https://devfeed.tech/tags/tracing.md>)

### AI overview

ResearchFlow-Eval assesses whether autoresearch systems can carry multiple interdependent claims through continuous research stages and revise experiments as evidence accumulates. It derives stage-specific rubrics from published anchor papers and scores work at both stage and claim levels. Across the evaluated general-purpose, skill-augmented, and research-orchestrated systems, experimental execution scored below design and planning. Systems often struggled to produce high-quality results that met claim requirements, and repeated attempts rarely improved scores. Skill augmentation and research-specific orchestration did not reliably raise performance. Results are averaged over two runs per system and scored by two LLM judges.

### Source excerpt

arXiv:2610.18088v3 Announce Type: replace-cross Abstract: Autoresearch systems increasingly undertake entire research projects. In practice, scientific research requires systems to pursue multiple interdependent claims through a continuous, multi-stage workflow and revise experiments as evidence accumulates. Existing evaluations of autoresearch systems, however, typically score isolated tasks or final artifacts, leaving these process-level capabilities largely unmeasured. We therefore introduce ResearchFlow-Eval, a fine-grained framework for assessing continuous scientific execution and iteration at both the stage and claim levels. Each evaluation is grounded in a published anchor paper, from which we derive stage-specific rubrics that decompose each research stage into dimensions and separately scored rubric items. Using these rubrics, we score each research stage and examine experimental execution in detail, focusing on multi-claim completion and revisions across attempts. We apply the evaluation framework to general-purpose agent harnesses, skill-augmented harnesses, and systems with research-specific orchestration. Averaged over two runs per system, each scored by two LLM judges, experimental execution lags behind design and planning for every evaluated system. Within execution, systems struggle to produce high-quality results and meet claim requirements, and repeated attempts rarely yield measurable improvements. Moreover, neither skill augmentation nor research-specific orchestration reliably yields higher scores. ResearchFlow-Eval thus extends autoresearch evaluation beyond isolated, single-objective runs to continuous workflows in which systems pursue multiple claims and refine their work across iterations.

## Google launches Gemini 4 Neon, touting lower costs for complex AI coding tasks

DevFeed: [Google launches Gemini 4 Neon, touting lower costs for complex AI coding tasks](<https://devfeed.tech/articles/google-launches-gemini-4-neon-touting-lower-costs-for-complex-ai-coding-tasks-143171.md>)

Original publisher: [Read original article](<https://www.infoworld.com/article/4250618/google-launches-gemini-4-neon-touting-lower-costs-for-complex-ai-coding-tasks.html>)

Author: Prasanth Aby Thomas

Published: 2026-11-19T09:49:58Z

Content type: news

Language: en

Sources: [InfoWorld](<https://devfeed.tech/sources/infoworld.md>)

Topics: [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [google-antigravity](<https://devfeed.tech/topics/google-antigravity.md>), [harvestbench](<https://devfeed.tech/topics/harvestbench.md>), [Test-driven development](<https://devfeed.tech/topics/tdd.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [antigravity](<https://devfeed.tech/tags/antigravity.md>), [api](<https://devfeed.tech/tags/api.md>), [api-pricing](<https://devfeed.tech/tags/api-pricing.md>), [artificial-analysis](<https://devfeed.tech/tags/artificial-analysis.md>), [artificial-analysis-intelligence-index](<https://devfeed.tech/tags/artificial-analysis-intelligence-index.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-development-tools-software-development](<https://devfeed.tech/tags/artificial-intelligence-development-tools-software-development.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [context-window](<https://devfeed.tech/tags/context-window.md>), [cost](<https://devfeed.tech/tags/cost.md>), [costs](<https://devfeed.tech/tags/costs.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [development-tools](<https://devfeed.tech/tags/development-tools.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [evaluations](<https://devfeed.tech/tags/evaluations.md>), [gemini](<https://devfeed.tech/tags/gemini.md>), [gemini-4](<https://devfeed.tech/tags/gemini-4.md>), [gemini-4-carbon](<https://devfeed.tech/tags/gemini-4-carbon.md>), [gemini-4-neon](<https://devfeed.tech/tags/gemini-4-neon.md>), [google](<https://devfeed.tech/tags/google.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [software-development](<https://devfeed.tech/tags/software-development.md>), [storytype-news](<https://devfeed.tech/tags/storytype-news.md>)

### AI overview

Google says Gemini 4 Neon can debug and refactor across large code repositories and, in Antigravity, coordinate parallel Gemini 4 Hydrogen subagents. Positioned between Gemini 4 Carbon for demanding projects and Hydrogen for speed, Neon offers four reasoning levels and a context window up to one million tokens. Its listed API prices are $3 per million input tokens, $15 per million output tokens, and $0.15 per million cached input tokens. Google reports 32% lower cost per task than Carbon in one evaluation and 50%-80% savings against competing models on two other benchmarks; these are company-reported results. The excerpt says enterprises should assess total cost per successful task, code quality, environment compatibility, productivity, and human intervention--not benchmark scores alone. It also notes production risks: generated code can be hard for developers to understand, agents may make conflicting changes, and reliability on complex or long-running tasks remains a concern. Experts describe supervised autonomy, developer oversight, and functional, end-to-end, and regression testing as important for production use.

### Source excerpt

Google is rolling out Gemini 4 Neon, a new model designed for complex software engineering tasks that the company says can complete work more economically than larger models, as it seeks to expand the use of AI coding agents among enterprise developers. Neon can work across large code repositories, performing multi-file debugging and refactoring while making targeted changes to existing code, according to Google. Within Google's Antigravity development environment, Neon can also orchestrate multiple Gemini 4 Hydrogen subagents working in parallel. Neon ranked first on the Artificial Analysis Coding Agent Index with a score of 58.6, while achieving 65.2% on Terminal-Bench 4.1, a benchmark used to evaluate AI agents performing tasks in terminal environments, Google said. Google is positioning Neon between Gemini 4 Carbon, designed for demanding projects, and Gemini 4 Hydrogen, which prioritizes faster execution. The company has set Neon's API pricing at $3 per million input tokens and $15 per million output tokens. Cached inputs cost $0.15 per million tokens, while the model supports a context window of up to one million tokens. Google said Neon delivered a 32% lower cost per task than Gemini 4 Carbon at the default high reasoning setting in Artificial Analysis Intelligence Index evaluations. It also reported savings of roughly 50% to 80% against leading competing models on the Vals Index and Agent Arena benchmarks. Neon offers four reasoning levels, allowing developers to adjust computational effort according to workload requirements. The model is being made available through the Gemini API in Google AI Studio and through Gemini for Business, with access also expanding to Google's Antigravity development platform. Lower costs could shift enterprise model choices Production software development often requires longer context windows, which can push token consumption beyond initial estimates, said Anand Joshi, managing director of market intelligence firm JP Data. This,

## Wayfinder: Frontier-level Code Search at a Fraction of the Cost

DevFeed: [Wayfinder: Frontier-level Code Search at a Fraction of the Cost](<https://devfeed.tech/articles/wayfinder-frontier-level-code-search-at-a-fraction-of-the-cost-141039.md>)

Original publisher: [Read original article](<https://blog.voyageai.com/2026/11/18/wayfinder-code-search/>)

Author: Voyage AI

Published: 2026-11-18T15:30:00Z

Content type: article

Language: en

Sources: [Voyage AI](<https://devfeed.tech/sources/voyage-ai.md>)

Topics: [code search](<https://devfeed.tech/topics/code-search.md>), [hybrid-search](<https://devfeed.tech/topics/hybrid-search.md>), [AI-driven information retrieval](<https://devfeed.tech/topics/ai-driven-information-retrieval.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Frontier AI](<https://devfeed.tech/topics/frontier-ai.md>), [Open Source Models & Datasets](<https://devfeed.tech/topics/open-source-models-datasets.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [code](<https://devfeed.tech/tags/code.md>), [code-search](<https://devfeed.tech/tags/code-search.md>), [cost](<https://devfeed.tech/tags/cost.md>), [frontier](<https://devfeed.tech/tags/frontier.md>), [grep](<https://devfeed.tech/tags/grep.md>), [news](<https://devfeed.tech/tags/news.md>), [search](<https://devfeed.tech/tags/search.md>), [wayfinder](<https://devfeed.tech/tags/wayfinder.md>)

### AI overview

Wayfinder pairs a small fine-tuned code-search model with a local hybrid index to make agent search cheaper while finding more relevant files. The index combines semantic search (80%) with BM25 (20%); its results are merged and reranked with grep results, then the model runs follow-up greps. On 200 held-out tasks, open models with this approach found about 74% of the right files for roughly $0.01 per search, compared with 76% for Sonnet 5.5 high effort at $0.12 and 78% for GPT-6.1 Sol medium effort at $0.07. Two pooled Wayfinder runs reached 78% for about $0.02. In a 42-task coding evaluation, competitive search methods had similar pass rates; Wayfinder reached 55%, while pooled runs reached 61%, but the sample was too small to establish a clear lead. Indexing also has ongoing costs: the authors estimate a typical repository pays back after about two searches monthly. Wayfinder is an open-source research project, not a MongoDB product; the authors recommend evaluating it on individual codebases.

### Source excerpt

Author: Henry Weller Before a coding agent can fix a bug, it has to find the code that matters. The coding agent industry has generally embraced grep as the way agents search code, but repeated searches with grep can become slow and expensive as codebases grow. For customers building coding agents, we wanted to explore [...]

## RoboTeam: Benchmarking Vision-Based Multi-Agent Control of Many Robot Arms in Real Time

DevFeed: [RoboTeam: Benchmarking Vision-Based Multi-Agent Control of Many Robot Arms in Real Time](<https://devfeed.tech/articles/roboteam-benchmarking-vision-based-multi-agent-control-of-many-robot-arms-in-real-time-139969.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22138>)

Author: Changxuan Fan, Zhongwei Xie, Boyi Zhang, Shaoli Hu, Chang Xu, Xiyu Ren, Yufei Li, Qiao Xiao, Huihao Jing, Kawai Chung, Wenbin Hu, Hongyu Luo, Lingyun Xie, Yisen Gao, Chunyang Li, Haoyu Huang, Wuganjin

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [Computer vision](<https://devfeed.tech/topics/computer-vision.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [real-time](<https://devfeed.tech/topics/real-time.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-control](<https://devfeed.tech/tags/agent-control.md>), [agents](<https://devfeed.tech/tags/agents.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [camera](<https://devfeed.tech/tags/camera.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ro](<https://devfeed.tech/tags/cs-ro.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [robot](<https://devfeed.tech/tags/robot.md>), [robot-arms](<https://devfeed.tech/tags/robot-arms.md>), [system](<https://devfeed.tech/tags/system.md>), [systems](<https://devfeed.tech/tags/systems.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

RoboTeam benchmarks agents controlling 4-16 robot arms across 36 long cooperative tasks, using camera images alone while simulated time continues during agent reasoning. In its adaptive setup, gpt-6.1-sol and opus-5.5 complete 28 and 27 tasks; four other tested models complete at most 3. Performance is costly and failure-prone: a typical gpt-6.1-sol task takes 3.5 hours and 218 million input tokens, with about a third of steps failing at least once and requiring retries. Four agent systems complete similar numbers of tasks but trade speed against token use: hierarchical and decentralized systems are faster and use up to 1.9 times the tokens, while single-agent uses fewer tokens but takes 1.6 times as long. The abstract concludes that no tested model or system is simultaneously fast, inexpensive, and reliable.

### Source excerpt

arXiv:2611.22138v1 Announce Type: cross Abstract: Multimodal agents can now control robot arms by calling tools and writing code. Existing benchmarks fall short in three ways: most test one or two arms on short tasks, many pause the world while the agent thinks, and many give it ground-truth object positions. We present RoboTeam, a simulated benchmark where agents control 4 to 16 robot arms in real time from camera images alone. Its 36 tasks are long jobs that the arms must do together, such as passing timbers or building a picture from hundreds of cubes. We start from the most flexible agent system, the adaptive system, where one agent decides when to fork delegates. With it, gpt-6.1-sol and opus-5.5 complete 28 and 27 tasks, and four other models at most 3. But at a high price: a typical task takes gpt-6.1-sol 3.5 hours and 218 million input tokens, and a third of its steps fail at least once and must be retried. We then compare it with three fixed systems: single-agent, hierarchical, and decentralized. All four complete a similar number of tasks but differ in time and tokens. The hierarchical and decentralized systems are faster but use up to 1.9 times the tokens, and the single agent uses fewer tokens but takes 1.6 times as long, so the adaptive system lies between them. Even so, no model or system is yet fast, cheap, and reliable at once. RoboTeam measures these limits where earlier benchmarks cannot: long multi-arm jobs, camera images only, and a world that never pauses.

## Multi-Agent Memory Inconsistency: Where Factual Conflicts Disappear in LLM Agents

DevFeed: [Multi-Agent Memory Inconsistency: Where Factual Conflicts Disappear in LLM Agents](<https://devfeed.tech/articles/multi-agent-memory-inconsistency-where-factual-conflicts-disappear-in-llm-agents-139989.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.22245>)

Author: Hejia Zhang, Zhongming Yu, Yimin Ding, Wei-Ting Chen, Hengjia Yu, Wentao Ni, Yijun Liu, Cong Xu, Xin Chen, Matrix Yao, Fan Zhao, Jishen Zhao

Published: 2026-11-18T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-delegation](<https://devfeed.tech/tags/agent-delegation.md>), [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [delegation](<https://devfeed.tech/tags/delegation.md>), [detection](<https://devfeed.tech/tags/detection.md>), [flow](<https://devfeed.tech/tags/flow.md>), [handoffs](<https://devfeed.tech/tags/handoffs.md>), [harvey](<https://devfeed.tech/tags/harvey.md>), [lead](<https://devfeed.tech/tags/lead.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llms](<https://devfeed.tech/tags/llms.md>), [models](<https://devfeed.tech/tags/models.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>)

### AI overview

In Harvey LAB document-review tasks, delegation improved retrieval of both conflicting claims but slightly reduced overall conflict detection. Pooled across three models at comparable generation spending, retrieval rose from 50.2% with single-agent execution to 68.7% with required delegation, while detection fell from 16.2% to 13.2%. Workers reported both claims in only 35.2% of retrieved pairs. Matched comparisons found detection was higher than single-agent execution when one worker read both claims, and lower when claims were split across workers. The results point to worker-to-lead handoffs as a key target for improving conflict detection; they are specific to this benchmark setting.

### Source excerpt

arXiv:2611.22245v1 Announce Type: new Abstract: Multi-agent delegation lets LLM systems distribute evidence collection across workers with separate contexts. This creates a consistency challenge: facts retrieved by workers may be missing from the evidence available to the lead for a joint conclusion. We introduce a benchmark and evidence-flow profiler for studying multi-agent memory inconsistency through controlled factual conflicts in Harvey LAB document-review tasks. Pooled across three models with comparable generation spending, requiring delegation increases retrieval of both conflicting claims from 50.2% to 68.7% relative to single-agent execution, yet conflict detection falls from 16.2% to 13.2%. Within this setting, workers report both claims for only 35.2% of retrieved pairs, and pooled matched comparisons show higher detection than single-agent execution when one worker reads both claims but lower detection when the claims are split across workers. Worker-to-lead handoffs are a central target for improving conflict detection in multi-agent LLMs.

## Perplexity Computer Adds Multiagent: Up to 16 AI Agents That Cross-Check Work

DevFeed: [Perplexity Computer Adds Multiagent: Up to 16 AI Agents That Cross-Check Work](<https://devfeed.tech/articles/perplexity-computer-adds-multiagent-up-to-16-ai-agents-that-cross-check-work-139437.md>)

Original publisher: [Read original article](<https://www.marktechpost.com/2026/11/17/perplexity-computer-multiagent-effort-level/>)

Author: Michal Sutter

Published: 2026-11-17T22:52:04Z

Content type: article

Language: en

Sources: [MarkTechPost](<https://devfeed.tech/sources/marktechpost.md>)

Topics: [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [supervisor agent pattern](<https://devfeed.tech/topics/supervisor-agent-pattern.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>)

Tags: [added](<https://devfeed.tech/tags/added.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-shorts](<https://devfeed.tech/tags/ai-shorts.md>), [applications](<https://devfeed.tech/tags/applications.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [claude](<https://devfeed.tech/tags/claude.md>), [computer](<https://devfeed.tech/tags/computer.md>), [cross](<https://devfeed.tech/tags/cross.md>), [editors-pick](<https://devfeed.tech/tags/editors-pick.md>), [effort](<https://devfeed.tech/tags/effort.md>), [for-devs](<https://devfeed.tech/tags/for-devs.md>), [gemini](<https://devfeed.tech/tags/gemini.md>), [gpt](<https://devfeed.tech/tags/gpt.md>), [harness](<https://devfeed.tech/tags/harness.md>), [lead](<https://devfeed.tech/tags/lead.md>), [models](<https://devfeed.tech/tags/models.md>), [multiagent](<https://devfeed.tech/tags/multiagent.md>), [new-releases](<https://devfeed.tech/tags/new-releases.md>), [other](<https://devfeed.tech/tags/other.md>), [perplexity](<https://devfeed.tech/tags/perplexity.md>), [perplexity-computer](<https://devfeed.tech/tags/perplexity-computer.md>), [post](<https://devfeed.tech/tags/post.md>), [share](<https://devfeed.tech/tags/share.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [staff](<https://devfeed.tech/tags/staff.md>), [tech-news](<https://devfeed.tech/tags/tech-news.md>), [technology](<https://devfeed.tech/tags/technology.md>), [uncategorized](<https://devfeed.tech/tags/uncategorized.md>), [up](<https://devfeed.tech/tags/up.md>), [work](<https://devfeed.tech/tags/work.md>)

### AI overview

Perplexity Computer's Multiagent effort level lets a lead agent handle part of a task and coordinate up to 16 supporting agents. Agents can share findings, challenge assumptions, flag discrepancies and edit shared files; users interact with the lead, which combines results and updates affected agents when work changes. The described default team uses Claude Opus 5.5, GPT 6.1 Sol and Gemini 4 Carbon, with other models available when requested or needed. Unlike Ultra, where supporting agents report only to the lead, Multiagent allows direct agent-to-agent collaboration; Model Council instead gathers independent answers for comparison. It is available on the web to Perplexity Max subscribers. The excerpt reports no benchmarks, accuracy or speed figures, context length, or per-task credit cost.

### Source excerpt

Perplexity added Multiagent to Computer, an effort level above Ultra where a lead agent runs up to 16 agents on Claude, GPT and Gemini models that share findings and check each other's work. The post Perplexity Computer Adds Multiagent: Up to 16 AI Agents That Cross-Check Work appeared first on MarkTechPost.

## Vibe synthesis: hierarchical agentic decision-making for intent-driven multistep molecule synthesis

DevFeed: [Vibe synthesis: hierarchical agentic decision-making for intent-driven multistep molecule synthesis](<https://devfeed.tech/articles/vibe-synthesis-hierarchical-agentic-decision-making-for-intent-driven-multistep-molecule-synthesis-137342.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.18772>)

Author: Jie Chen, Guikun Chen, Feng Wan, Ning Xu, Wen Liu, Yixiang Ruan, Yiyi Qiu, Haohua Li, Wei Wang, Yi Wang, Xiaoxue Wang, Peng Wang, Jiazhou Zhu, Jun Xuan, Chengxi Li, Jianhong Xu, Wenguan Wang, Yiming M

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Multi-step AI processes](<https://devfeed.tech/topics/multi-step-ai-processes.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [decision-making](<https://devfeed.tech/topics/decision-making.md>)

Tags: [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [amino-acids](<https://devfeed.tech/tags/amino-acids.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [automation](<https://devfeed.tech/tags/automation.md>), [autonomy](<https://devfeed.tech/tags/autonomy.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [decision](<https://devfeed.tech/tags/decision.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [horizon](<https://devfeed.tech/tags/horizon.md>), [laboratory](<https://devfeed.tech/tags/laboratory.md>), [laboratory-automation](<https://devfeed.tech/tags/laboratory-automation.md>), [medicine](<https://devfeed.tech/tags/medicine.md>), [physics-chem-ph](<https://devfeed.tech/tags/physics-chem-ph.md>), [synthesis](<https://devfeed.tech/tags/synthesis.md>), [target](<https://devfeed.tech/tags/target.md>), [workflow](<https://devfeed.tech/tags/workflow.md>)

### AI overview

VibeSyn uses hierarchical agents to turn chemist-defined molecular goals and accumulating experimental evidence into adaptive synthesis decisions, executed with computational tools and laboratory automation. The authors report three demonstrations, ranging from evidence-grounded experimental decisions to a ten-step total synthesis of catharanthine and synthesis of cyclic peptides containing unnatural amino acids. In the long-horizon campaigns, active human input and simple confirmation represented less than 7.5% and 5.2%, respectively, of recorded workflow actions and outcomes. The supplied text is an abstract and does not provide further methodological or performance details.

### Source excerpt

arXiv:2611.18772v1 Announce Type: cross Abstract: The ability to construct complex molecules underpins advances in medicine and the molecular sciences. Yet long-horizon multistep synthesis requires decisions to be revised as evidence accumulates across chemically dependent operations, a capability not readily supported by artificial-intelligence tools and automated platforms organized around predefined tasks and hard-coded workflows. Here we propose and demonstrate a vibe synthesis system (VibeSyn), an intent-driven architecture in which hierarchical agents translate chemist-defined molecular objectives and accumulating evidence into adaptive decisions executed through computational tools and laboratory automation. We evaluate VibeSyn in three demonstrations of increasing scope and autonomy, progressing from evidence-grounded experimental decisions to adaptive execution of a ten-step total synthesis of catharanthine and structure-to-compound synthesis of cyclic peptides containing unnatural amino acids (uAAs). Across the long-horizon campaigns, active human input and simple confirmation account for less than 7.5% and 5.2%, respectively, of recorded workflow actions and outcomes. VibeSyn points towards a future in which chemists define synthetic intent and agent-directed laboratories autonomously navigate the evolving experimental trajectory required to deliver the target compound.

## Autonomous Multi-modal Agents for ITOps Observability

DevFeed: [Autonomous Multi-modal Agents for ITOps Observability](<https://devfeed.tech/articles/autonomous-multi-modal-agents-for-itops-observability-137520.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19982>)

Author: Gunjan Siddharth, Sanjana Garai, Utsav Shekhar, Prakhar Jain, Sarvam Maheshwari, Balaji Ganesan, Avirup Saha, Seema Nagar, Pratik Mishra, Felix George, Suranjana Samanta, Harshit Kumar, Mudit Verma, P

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agent observability](<https://devfeed.tech/topics/agent-observability.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [GraphQL](<https://devfeed.tech/topics/graphql.md>), [LLM observability](<https://devfeed.tech/topics/llm-observability.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [defects](<https://devfeed.tech/tags/defects.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [graphql](<https://devfeed.tech/tags/graphql.md>), [itops](<https://devfeed.tech/tags/itops.md>), [lifecycle](<https://devfeed.tech/tags/lifecycle.md>), [llm](<https://devfeed.tech/tags/llm.md>), [mapping](<https://devfeed.tech/tags/mapping.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [natural-language](<https://devfeed.tech/tags/natural-language.md>), [observability](<https://devfeed.tech/tags/observability.md>), [operations](<https://devfeed.tech/tags/operations.md>), [routing](<https://devfeed.tech/tags/routing.md>), [schema](<https://devfeed.tech/tags/schema.md>), [serving](<https://devfeed.tech/tags/serving.md>), [studio](<https://devfeed.tech/tags/studio.md>), [system](<https://devfeed.tech/tags/system.md>), [systems](<https://devfeed.tech/tags/systems.md>), [tool](<https://devfeed.tech/tags/tool.md>)

### AI overview

ITOps Observability Studio coordinates specialized LLM agents to connect observability tools and automate workflows, including natural-language-to-GraphQL queries, schema generation and mapping, trace data lifecycle management, and visualization. Its three-tier design supports inter-agent communication and unified tool integration, with analytics tracing agent model and tool calls. Agents are evaluated separately using gold answers where available and model judging for schema mapping and chart quality. The abstract reports strong query generation on a public benchmark and routing with a current open model, and success in most data lifecycle scenarios. It also says evaluation exposed defects arising from interactions among prompts, parsers, and serving stacks, along with fixes; no numerical results are given.

### Source excerpt

arXiv:2611.19982v1 Announce Type: new Abstract: ITOps observability products must integrate data from different tools and systems. We present ITOps Observability Studio, a multi-modal multi-agent system that orchestrates observability workflows through collaborative AI agents. Its three-tier architecture features inter-agent communication, specialized LLM-powered agents for data access, schema operations and visualization, and unified tool integration. The system supports natural language to GraphQL translation, GraphQL schema generation, schema mapping, trace data lifecycle management and multi-tool workflows for reliability engineering. Its agents are instrumented with agent analytics, so that their model and tool calls can be traced. We evaluate each agent on its own benchmark, scored against gold answers where they exist and by a model judge for schema mapping and chart quality. Query generation on a public benchmark and routing work well with a current open model, and the data lifecycle management agent solves most of its scenarios. The evaluation exposed defects that only appear when prompts, parsers and serving stacks are tested together, and we report them together with their fixes.

## Multi-LLM Agentic AI at the Edge: Joint Agent Scheduling and Resource Allocation

DevFeed: [Multi-LLM Agentic AI at the Edge: Joint Agent Scheduling and Resource Allocation](<https://devfeed.tech/articles/multi-llm-agentic-ai-at-the-edge-joint-agent-scheduling-and-resource-allocation-137630.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.20915>)

Author: Yuanyan Song, Kezhi Wang, Xinmian Xu, Yongmin Li

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [supervisor agent pattern](<https://devfeed.tech/topics/supervisor-agent-pattern.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Operations research and optimization](<https://devfeed.tech/topics/operations-research-and-optimization.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [scheduling](<https://devfeed.tech/topics/scheduling.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-ai-architecture](<https://devfeed.tech/tags/agentic-ai-architecture.md>), [agentic-workflows](<https://devfeed.tech/tags/agentic-workflows.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-architecture](<https://devfeed.tech/tags/ai-architecture.md>), [ai-at-the-edge](<https://devfeed.tech/tags/ai-at-the-edge.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [approximation](<https://devfeed.tech/tags/approximation.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [bandwidth](<https://devfeed.tech/tags/bandwidth.md>), [convex](<https://devfeed.tech/tags/convex.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [edge](<https://devfeed.tech/tags/edge.md>), [energy](<https://devfeed.tech/tags/energy.md>), [llm](<https://devfeed.tech/tags/llm.md>), [resource](<https://devfeed.tech/tags/resource.md>), [scheduling](<https://devfeed.tech/tags/scheduling.md>)

### AI overview

The paper proposes an edge architecture in which a supervisor splits user requests into subtasks and dispatches them to worker agents through parallel or serial workflows, with a two-level retry mechanism. To reduce energy use while meeting latency and QoS constraints, an alternating-optimization scheme jointly handles agent selection and resource allocation. The parallel workflow allocates bandwidth; the serial workflow allocates service time using KKT conditions. Both address retrieval and agent selection, using SCA, Lagrangian relaxation, and Gale-Shapley or LLM-based algorithms. Simulations report lower energy consumption than benchmark schemes, and experiments using Qwen and Llama report improved task success rates.

### Source excerpt

arXiv:2611.20915v1 Announce Type: new Abstract: In this paper, an edge-enabled multi-large language model (LLM) agentic AI architecture is proposed, where one supervisor AI agent and multiple worker AI agents cooperatively execute users' requests at the edge. The supervisor AI agent decouples the user request into a series of sub-tasks and dynamically dispatches them to LLM-equipped worker AI agents via two agentic workflows, i.e., parallel and serial workflows, with a two-level retry mechanism to enhance execution reliability. To prolong the service duration of worker AI agents, the energy minimization problems are formulated by jointly considering AI agent selection and resource allocation, subject to task execution latency and quality of service (QoS) constraints. A resource allocation scheme based on alternating optimization (AO) is proposed. Specifically, for the parallel workflow, the scheme decouples the original problem into bandwidth allocation, retrieval selection and AI agent selection sub-problems, which are iteratively solved by the proposed successive convex approximation (SCA), Lagrangian relaxation (LR), and Gale-Shapley (GS) or LLM-based algorithms, respectively. For the serial workflow, the problem is similarly decoupled into service time allocation, retrieval selection and AI agent selection sub-problems; the optimal service time allocation is obtained via Karush-Kuhn-Tucker (KKT) conditions, while the latter two sub-problems are addressed by extending the proposed algorithms to the serial scenario. Simulation results show that the proposed resource allocation scheme can reduce system energy consumption compared with other benchmarks, while experiments on a multi-LLM agentic AI system based on Qwen and Llama demonstrate improved task execution success rates.

## Advancing ESG Intelligence: An Expert-level Agent and Comprehensive Benchmark for Sustainable Finance

DevFeed: [Advancing ESG Intelligence: An Expert-level Agent and Comprehensive Benchmark for Sustainable Finance](<https://devfeed.tech/articles/advancing-esg-intelligence-an-expert-level-agent-and-comprehensive-benchmark-for-sustainable-finance-137236.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2601.08676>)

Author: Yilei Zhao, Wentao Zhang, Lei Xiao, Yandan Zheng, Mengpu Liu, Wei Yang Bryan Lim

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [atomic](<https://devfeed.tech/tags/atomic.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [charts](<https://devfeed.tech/tags/charts.md>), [corporate-sustainability](<https://devfeed.tech/tags/corporate-sustainability.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [data](<https://devfeed.tech/tags/data.md>), [data-fragmentation](<https://devfeed.tech/tags/data-fragmentation.md>), [domain](<https://devfeed.tech/tags/domain.md>), [esg](<https://devfeed.tech/tags/esg.md>), [evaluations](<https://devfeed.tech/tags/evaluations.md>), [finance](<https://devfeed.tech/tags/finance.md>), [general](<https://devfeed.tech/tags/general.md>), [generate](<https://devfeed.tech/tags/generate.md>), [governance](<https://devfeed.tech/tags/governance.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>)

### AI overview

ESGAgent is a hierarchical multi-agent system that combines retrieval augmentation, web search, and domain-specific functions to support in-depth ESG analysis amid fragmented unstructured data and complex auditing workflows. The paper also introduces a three-level benchmark based on 310 corporate sustainability reports, spanning atomic questions through integrated analysis. In reported evaluations, ESGAgent achieved 84.15% average accuracy on atomic question-answering tasks, outperforming state-of-the-art closed-source LLMs, and generated professional reports with charts and verifiable references. The supplied abstract does not detail the benchmark levels or comparative evaluation methods.

### Source excerpt

arXiv:2601.08676v3 Announce Type: replace Abstract: Environmental, social, and governance (ESG) criteria are essential for evaluating corporate sustainability and ethical performance. However, professional ESG analysis is hindered by data fragmentation across unstructured sources, and existing large language models (LLMs) often struggle with the complex, multi-step workflows required for rigorous auditing. To address these limitations, we introduce ESGAgent, a hierarchical multi-agent system empowered by a specialized toolset, including retrieval augmentation, web search and domain-specific functions, to generate in-depth ESG analysis. Complementing this agentic system, we present a comprehensive three-level benchmark derived from 310 corporate sustainability reports, designed to evaluate capabilities ranging from atomic common-sense questions to the generation of integrated, in-depth analysis. Empirical evaluations demonstrate that ESGAgent outperforms state-of-the-art closed-source LLMs with an average accuracy of 84.15% on atomic question-answering tasks, and excels in professional report generation by integrating rich charts and verifiable references. These findings confirm the diagnostic value of our benchmark, establishing it as a vital testbed for assessing general and advanced agentic capabilities in high-stakes vertical domains.

## Assume/Guarantee Contracts for the Compositional Multi-Agent Reinforcement Learning of Heterogeneous Teams

DevFeed: [Assume/Guarantee Contracts for the Compositional Multi-Agent Reinforcement Learning of Heterogeneous Teams](<https://devfeed.tech/articles/assume-guarantee-contracts-for-the-compositional-multi-agent-reinforcement-learning-of-heterogeneous-teams-137389.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.19078>)

Author: Dominic Klukas, Cyrus Neary

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>)

Tags: [algorithm](<https://devfeed.tech/tags/algorithm.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [contracts](<https://devfeed.tech/tags/contracts.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [learning](<https://devfeed.tech/tags/learning.md>), [low-level](<https://devfeed.tech/tags/low-level.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [multi-agent-systems](<https://devfeed.tech/tags/multi-agent-systems.md>), [probability](<https://devfeed.tech/tags/probability.md>), [progress](<https://devfeed.tech/tags/progress.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [reuse](<https://devfeed.tech/tags/reuse.md>), [robust](<https://devfeed.tech/tags/robust.md>), [synchronization](<https://devfeed.tech/tags/synchronization.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [team](<https://devfeed.tech/tags/team.md>), [teams](<https://devfeed.tech/tags/teams.md>), [theoretical](<https://devfeed.tech/tags/theoretical.md>), [time](<https://devfeed.tech/tags/time.md>), [timing](<https://devfeed.tech/tags/timing.md>), [training](<https://devfeed.tech/tags/training.md>)

### AI overview

The paper proposes AGC-MAS, compositional contracts that specify a subtask, progress-tracking automaton, timeout and success-probability requirements, and dispatch behavior. These contracts let heterogeneous agents substitute for one another and allow previously satisfied contracts to be reused. IHA-MARL decomposes team tasks into contracts, then iterates between assigning contracts to agents, training them, and assessing their capabilities until the desired team success probability is met. Across three benchmarks with five seeds each, it reports 100% observed final evaluation success; CPU timing pilots found comparable baselines required about 7-11 times more training-loop time. These results are reported in the abstract.

### Source excerpt

arXiv:2611.19078v1 Announce Type: new Abstract: Multi-agent reinforcement learning (MARL) problems for heterogeneous teams require agents to collaborate to complete team tasks while making the best use of diverse agent capabilities. We introduce assume/guarantee contracts for heterogeneous multi-agent systems (AGC-MAS) and Iterative Heterogeneous Assignment for MARL (IHA-MARL). AGC-MAS contracts are compositional objects that encode the assumptions provided to, and the guarantees required of, each subtask contributing to the overall team task. We formulate contracts as tuples that specify a subtask, an automaton, timeout and success-probability requirements, and dispatch behavior. The automaton tracks an agent's high-level progress through the subtask associated with its contract and provides a natural mechanism for synchronization across contracts. Our abstraction enables robust interaction across heterogeneous low-level subtask implementations, allows heterogeneous agents to substitute for one another in satisfying contracts, and supports reuse of previously satisfied contracts in new team contexts. Having defined our theoretical framework, we develop a decomposition algorithm that breaks down a team task into a set of contracts. We then propose an iterative algorithm for task assignment and MARL that alternates between assigning contracts representing subtasks to individual agents, training the assigned agents to satisfy their contracts, and assessing their resulting capabilities to inform the next round of task assignments until the team's success probability meets the desired threshold. Our experimental results demonstrate the versatility of contracts in enabling agent substitution and contract reuse. Across three benchmarks with five seeds each, IHA-MARL achieves observed $100\%$ final evaluation success. In CPU timing pilots, comparable baselines required approximately $7$--$11\times$ more training-loop time.

## WeVisForensics: Self-Evolving Agent for Omni AI-Generated Video Detection

DevFeed: [WeVisForensics: Self-Evolving Agent for Omni AI-Generated Video Detection](<https://devfeed.tech/articles/wevisforensics-self-evolving-agent-for-omni-ai-generated-video-detection-137688.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.21373>)

Author: Yipeng Du, Yuxi Li, Ying Zhang, Chong Sun, Chen Li, Jing Lyu, Ying Tai

Published: 2026-11-17T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [supervisor agent pattern](<https://devfeed.tech/topics/supervisor-agent-pattern.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Video Analytics](<https://devfeed.tech/topics/video-analytics.md>), [multimodal-ai](<https://devfeed.tech/topics/multimodal-ai.md>), [image recognition](<https://devfeed.tech/topics/image-recognition.md>)

Tags: [abstract](<https://devfeed.tech/tags/abstract.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-workflows](<https://devfeed.tech/tags/agent-workflows.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-generated-video](<https://devfeed.tech/tags/ai-generated-video.md>), [annotations](<https://devfeed.tech/tags/annotations.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [audio](<https://devfeed.tech/tags/audio.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [call](<https://devfeed.tech/tags/call.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [chain-of-thought](<https://devfeed.tech/tags/chain-of-thought.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-cv](<https://devfeed.tech/tags/cs-cv.md>), [detection](<https://devfeed.tech/tags/detection.md>), [domain](<https://devfeed.tech/tags/domain.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [video](<https://devfeed.tech/tags/video.md>)

### AI overview

WeVisForensics detects AI-generated videos by coordinating a main investigator with visual and audio forensic inspector agents. The investigator selects which artifact to inspect, the time window, and when to stop; shared feedback also helps refine the inspectors. Its dual-timescale policy optimization trains the investigator faster and improves inspectors more gradually using annotated artifact intervals. ForenCue-30K provides role-specific annotations, interaction trajectories, and a reinforcement-learning pool. The abstract reports 95.2% average detection accuracy across in-domain, out-of-domain, and agentic benchmarks, 9.0% above the strongest prior detector, with verdicts grounded in localized evidence.

### Source excerpt

arXiv:2611.21373v1 Announce Type: cross Abstract: Recent advances in video generation call for reliable detection grounded in explicit forensic evidence. However, current methods primarily depend on single-pass chain-of-thought reasoning that overlooks brief or localized artifacts, or on predefined multi-agent workflows that offer limited flexibility in selecting relevant inspections. Even with adaptive investigation, optimizing the investigation policy alone does not directly improve the forensic capabilities of fixed experts. We introduce WeVisForensics, an agentic framework in which a main investigator agent coordinates a self-evolving library of forensic inspector sub-agents spanning visual and audio cues. The investigator learns which artifact type to examine, over which window, and when to conclude, and these interactions provide shared feedback for both levels of agents. Our dual-timescale policy optimization strategy, DTPO, closes this loop through fast reinforcement learning of the investigator and slower refinement of the inspectors, guided by annotated artifact intervals, to report subtle artifacts they once dismissed. Furthermore, we curate ForenCue-30K, combining role-specific forensic annotations and interaction trajectories to supervise individual inspections and their coordination, with an accompanying reinforcement-learning pool. Experiments show that WeVisForensics averages 95.2% detection accuracy across in-domain, out-of-domain and agentic benchmarks, 9.0% above the strongest prior detector, while grounding its verdicts in localized evidence.

## Taxonomy of multi-agent design patterns, anti-patterns, and lessons from manufacturing

DevFeed: [Taxonomy of multi-agent design patterns, anti-patterns, and lessons from manufacturing](<https://devfeed.tech/articles/taxonomy-of-multi-agent-design-patterns-anti-patterns-and-lessons-from-manufacturing-136364.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/taxonomy-of-multi-agent-design-patterns-anti-patterns-and-lessons-from-manufacturing/>)

Author: Malini Sethi

Published: 2026-11-16T21:53:46Z

Content type: article

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [supervisor agent pattern](<https://devfeed.tech/topics/supervisor-agent-pattern.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [Multi-step AI processes](<https://devfeed.tech/topics/multi-step-ai-processes.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Physical AI](<https://devfeed.tech/topics/physical-ai.md>), [Parallel reasoning](<https://devfeed.tech/topics/parallel-reasoning.md>)

Tags: [advanced-300](<https://devfeed.tech/tags/advanced-300.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-loops](<https://devfeed.tech/tags/agent-loops.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [amazon-bedrock](<https://devfeed.tech/tags/amazon-bedrock.md>), [amazon-bedrock-guardrails](<https://devfeed.tech/tags/amazon-bedrock-guardrails.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [automotive](<https://devfeed.tech/tags/automotive.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [aws](<https://devfeed.tech/tags/aws.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [cost](<https://devfeed.tech/tags/cost.md>), [design](<https://devfeed.tech/tags/design.md>), [design-patterns](<https://devfeed.tech/tags/design-patterns.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-systems](<https://devfeed.tech/tags/distributed-systems.md>), [framework](<https://devfeed.tech/tags/framework.md>), [manufacturing](<https://devfeed.tech/tags/manufacturing.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [taxonomy](<https://devfeed.tech/tags/taxonomy.md>)

### AI overview

The post recommends starting with deterministic Tier 1 workflows and using autonomous multi-agent coordination only when subtasks emerge during the work, since flexibility adds token cost, latency, and debugging burden. It distinguishes pipelines, conditional workflows, routers, and parallel workflows from plan-based orchestration, handoffs, shared conversations, and debate. It advises hard retry and token limits, schema-based handoffs, bounded agents, and approval gates for actions affecting customers or equipment. For manufacturing, it emphasizes p99 latency budgets and deterministic fallbacks, keeping PLC/SCADA/MES control deterministic, validating recommendations against current OT state before approved actuation, processing time-sensitive sensor data at the edge, and structuring agent memory and handoffs around shifts. AWS IoT Greengrass is named for edge monitoring; the supplied text does not detail the promised broader service mapping.

### Source excerpt

Multi-agent coordination is a distributed systems problem, not a model problem. This post gives a field-tested taxonomy of multi-agent design patterns and anti-patterns, a framework for deciding when multiple agents are warranted, and how to build cost-efficient, reliable systems on AWS.

## Only as Safe as the Weakest Agent: A Break That Appears Only When the System Is Multi-Agent

DevFeed: [Only as Safe as the Weakest Agent: A Break That Appears Only When the System Is Multi-Agent](<https://devfeed.tech/articles/only-as-safe-as-the-weakest-agent-a-break-that-appears-only-when-the-system-is-multi-agent-135074.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17520>)

Author: Keegan Wang, Anantika Mannby

Published: 2026-11-16T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [ai-safety](<https://devfeed.tech/tags/ai-safety.md>), [attack](<https://devfeed.tech/tags/attack.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [codex](<https://devfeed.tech/tags/codex.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [delegates](<https://devfeed.tech/tags/delegates.md>), [delivery](<https://devfeed.tech/tags/delivery.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [multi-agent-systems](<https://devfeed.tech/tags/multi-agent-systems.md>), [orchestrator](<https://devfeed.tech/tags/orchestrator.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [repeat](<https://devfeed.tech/tags/repeat.md>)

### AI overview

The study reports that multi-agent systems can fail on planted forged records even when their orchestrator model is robust when searching the same files alone. In tests using unmodified Claude Code, Codex, and a reference harness, delegating to cheaper sub-agents led to harmful revisions in 20/180, 44/180, 75/180, and 553/900 trials; direct orchestrator searches produced almost zero in more than 1,000 trials. Simulations across 18 models found five aggregators repeated a corrupted sub-agent's error in 751/757 trials, despite being almost never unsafe on their own. The authors conclude that system safety can track the weaker agent's behavior rather than the orchestrator's. These are reported experimental results; the supplied text does not specify which rate belongs to which system or sub-agent configuration.

### Source excerpt

arXiv:2611.17520v1 Announce Type: cross Abstract: Major industry multi-agent systems like Codex and Claude Code can be broken by attacks that have long been unsuccessful against modern frontier models. AI Safety as a discipline has largely focused on finding the bounds and vulnerabilities of AI models used in isolation. Yet as humans increasingly deploy these models as agents and multi-agent systems, the same attacks that had been thought to be patched when using a model in isolation can break these same models under specific circumstances. We show that these system breaks are not purely theoretical, but can break Claude Code, Codex, and a reference harness, all unmodified and used with their own sub-agent mechanisms. We show the attack by asking agents to answer questions from a store of files on disk in which we have planted a forged record. Searching the store themselves, the strong model orchestrators commit almost zero harmful revisions in over 1,000 trials. However, if that same orchestrator model delegates to a cheaper sub-agent, which is what these harnesses are built to do, the same models on the same facts fail on 20/180, 44/180, 75/180 and 553/900. Thus, even when the system is driven by a strong model, the system's safety lands near the weak reader's, not the orchestrator's. Simulations over 18 models show the same effect in the delivery channel where five aggregators, who on their own were almost never unsafe, repeat a corrupted sub-agent's error in 751/757 trials. The safety of real-world agentic systems that use strong orchestrator agents and weaker sub-agents, and multi-agent systems in general, is defined not by the safety of the strong agent but by the safety of the weakest agent.

## The Agent That Polled 55 Times to Learn Nothing

DevFeed: [The Agent That Polled 55 Times to Learn Nothing](<https://devfeed.tech/articles/the-agent-that-polled-55-times-to-learn-nothing-133912.md>)

Original publisher: [Read original article](<https://hackernoon.com/the-agent-that-polled-55-times-to-learn-nothing?source=rss>)

Author: Dmitriy Fedoryshchev

Published: 2026-11-14T00:50:08Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [cache](<https://devfeed.tech/tags/cache.md>), [caching](<https://devfeed.tech/tags/caching.md>), [change](<https://devfeed.tech/tags/change.md>), [claude](<https://devfeed.tech/tags/claude.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cron](<https://devfeed.tech/tags/cron.md>), [developer-tools](<https://devfeed.tech/tags/developer-tools.md>), [llms](<https://devfeed.tech/tags/llms.md>), [multi-agebts](<https://devfeed.tech/tags/multi-agebts.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [setup](<https://devfeed.tech/tags/setup.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [token](<https://devfeed.tech/tags/token.md>), [tokens](<https://devfeed.tech/tags/tokens.md>)

### AI overview

In a three-layer agent pipeline, the Lead spent 3.66 of 19.3 hours sleeping and made 55 unnecessary long-sleep-then-poll cycles in 303 tool calls. Those turns processed about 13.5 million tokens, mostly repeated cached context; the author estimates five checks would have used about 1.2 million. Sleeps also blocked the Lead from responding to human corrections. The author flattened the setup to work with the Lead directly, required a stated reason for sleeps, and used cron-style reminders for team check-ins; implementers could still wait on self-terminating tasks. Over two weeks, 75 subsequent calls by that Lead and ten later Leads had no wasted sleeps. Small-fix median time fell from 23.2 to 5.6 minutes, though the fast lane contributed. The author cannot establish pipeline-wide token savings, and cautions that the flatter setup reflects a personal preference, not a universal recommendation.

### Source excerpt

Why a lead agent in a multi-agent pipeline kept sleeping, polling and ignoring its human, what it cost in tokens, and how a flatter setup fixed it.

## From Delegation to Coordination: Learning Efficient Multi-Agent Deep Research

DevFeed: [From Delegation to Coordination: Learning Efficient Multi-Agent Deep Research](<https://devfeed.tech/articles/from-delegation-to-coordination-learning-efficient-multi-agent-deep-research-132305.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.17041>)

Author: Yuyang Wu, Yang Liu, Zequn Xie, Junjie Wang, Dan Yang, Zehao Tan, Ding-Chu Zhang, Junming Yang, Yue Shen, Jian Wang, Jinjie Gu

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [Reinforcement learning](<https://devfeed.tech/topics/reinforcement-learning.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-delegation](<https://devfeed.tech/tags/agent-delegation.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [context](<https://devfeed.tech/tags/context.md>), [contribution](<https://devfeed.tech/tags/contribution.md>), [cost](<https://devfeed.tech/tags/cost.md>), [credit-assignment](<https://devfeed.tech/tags/credit-assignment.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-ma](<https://devfeed.tech/tags/cs-ma.md>), [deep-research](<https://devfeed.tech/tags/deep-research.md>), [delegation](<https://devfeed.tech/tags/delegation.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [execution](<https://devfeed.tech/tags/execution.md>), [framework](<https://devfeed.tech/tags/framework.md>), [learning](<https://devfeed.tech/tags/learning.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [research](<https://devfeed.tech/tags/research.md>), [subagent](<https://devfeed.tech/tags/subagent.md>), [subagents](<https://devfeed.tech/tags/subagents.md>), [systems](<https://devfeed.tech/tags/systems.md>), [team](<https://devfeed.tech/tags/team.md>)

### AI overview

LEAD-RL jointly trains the main research agent and its subagents, using a training-time rubric to assess each subagent report's research utility and gate its share of the team advantage. Final-answer correctness remains the sole team reward; the method adds no token, step, or speedup reward. Across five benchmarks and two model families, it improves answer quality. On BrowseComp, it raises accuracy by 2.49 percentage points and cuts critical-path cost by 24.5% versus joint training with shared team credit. In validation, accuracy rises by 6.78 points at approximately constant critical-path cost.

### Source excerpt

arXiv:2611.17041v1 Announce Type: cross Abstract: Multi-agent deep-research systems use delegation to parallelize search and keep long tool traces out of the main agent's context. Yet delegation helps only when task allocation and subtask execution are coordinated: the main agent must issue useful subtasks, while subagents must return reliable evidence. Existing training either optimizes only the main agent or broadcasts the same team advantage to all agents, which can reward misleading subagent reports and penalize useful ones. We introduce LEAD-RL (Learning Effective Agent Delegation with Reinforcement Learning), a joint main-agent-subagent training framework with contribution-aware credit assignment. A training-time rubric evaluates the research utility of each subagent report and gates its inherited team advantage, while final-answer correctness remains the only team reward; no token, step, or speedup reward is used. Across five benchmarks and two model families, LEAD-RL improves answer quality. On BrowseComp, it improves accuracy by 2.49 percentage points and reduces critical-path cost by 24.5% over joint training with shared team credit. During validation, accuracy increases by 6.78 points at approximately constant critical-path cost. These results show that contribution-aware credit assignment improves the quality-efficiency trade-off of multi-agent deep research.

## How DXC Technology built a multi-model AI substrate on Amazon Bedrock

DevFeed: [How DXC Technology built a multi-model AI substrate on Amazon Bedrock](<https://devfeed.tech/articles/how-dxc-technology-built-a-multi-model-ai-substrate-on-amazon-bedrock-129623.md>)

Original publisher: [Read original article](<https://aws.amazon.com/blogs/machine-learning/how-dxc-technology-built-a-multi-model-ai-substrate-on-amazon-bedrock/>)

Author: Asif Fouzi

Published: 2026-11-11T20:45:45Z

Content type: article

Language: en

Sources: [Artificial Intelligence](<https://devfeed.tech/sources/artificial-intelligence.md>)

Topics: [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [Amazon Bedrock Guardrails](<https://devfeed.tech/topics/amazon-bedrock-guardrails.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Amazon Bedrock Knowledge Bases](<https://devfeed.tech/topics/amazon-bedrock-knowledge-bases.md>), [Amazon EventBridge](<https://devfeed.tech/topics/amazon-eventbridge.md>), [VPC Flow Logs](<https://devfeed.tech/topics/vpc-flow-logs.md>), [Amazon CloudWatch](<https://devfeed.tech/topics/amazon-cloudwatch.md>), [Managed Services](<https://devfeed.tech/topics/managed-services.md>), [amazon](<https://devfeed.tech/topics/amazon.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [amazon](<https://devfeed.tech/tags/amazon.md>), [amazon-bedrock](<https://devfeed.tech/tags/amazon-bedrock.md>), [audits](<https://devfeed.tech/tags/audits.md>), [bedrock](<https://devfeed.tech/tags/bedrock.md>), [critical](<https://devfeed.tech/tags/critical.md>), [customer-solutions](<https://devfeed.tech/tags/customer-solutions.md>), [harness](<https://devfeed.tech/tags/harness.md>), [intermediate-200](<https://devfeed.tech/tags/intermediate-200.md>), [it](<https://devfeed.tech/tags/it.md>), [it-services](<https://devfeed.tech/tags/it-services.md>), [learn](<https://devfeed.tech/tags/learn.md>), [managed](<https://devfeed.tech/tags/managed.md>), [mission](<https://devfeed.tech/tags/mission.md>), [model](<https://devfeed.tech/tags/model.md>), [oasis](<https://devfeed.tech/tags/oasis.md>), [organizations](<https://devfeed.tech/tags/organizations.md>), [scale](<https://devfeed.tech/tags/scale.md>), [substrate](<https://devfeed.tech/tags/substrate.md>), [technology](<https://devfeed.tech/tags/technology.md>)

### AI overview

DXC separates DXC OASIS into a governed model substrate on Amazon Bedrock and a DXC-built harness for orchestration, operational context, policy checks, human approval, and audit trails. Bedrock gives the system one API for Claude, Nova, Llama, and open-weight models, with Guardrails for content and PII controls, PrivateLink for private inference traffic, and regional controls for data residency. The shared API enables workload-based model routing, but the harness evaluates models against task suites before production use because models differ in behavior. Agents start read-only; production changes require recorded approval in the described workflow. DXC reports a 95% average reduction in resolution time across 56 measured use cases and diagnostic accuracy of at least 92%, validated against senior engineers.

### Source excerpt

DXC Technology runs mission-critical managed IT services for regulated organizations across 70 countries. Learn how DXC built DCX OASIS, its Human+AI operating model, on a governed multi-model substrate in Amazon Bedrock, separating a resilient model substrate from a trusted harness that orchestrates, governs, and audits AI agents at scale.

## Anthropic unveils Claude Fable 5.2, Mythos 5.2 with coding gains and lower prices

DevFeed: [Anthropic unveils Claude Fable 5.2, Mythos 5.2 with coding gains and lower prices](<https://devfeed.tech/articles/anthropic-unveils-claude-fable-5-2-mythos-5-2-with-coding-gains-and-lower-prices-125646.md>)

Original publisher: [Read original article](<https://www.computerworld.com/article/4246693/anthropic-unveils-claude-fable-5-2-mythos-5-2-with-coding-gains-and-lower-prices-2.html>)

Author: Prasanth Aby Thomas

Published: 2026-11-10T09:35:38Z

Content type: news

Language: en

Sources: [Computerworld](<https://devfeed.tech/sources/computerworld.md>)

Topics: [Claude](<https://devfeed.tech/topics/claude.md>), [anthropic](<https://devfeed.tech/topics/anthropic.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>), [snyk-platform](<https://devfeed.tech/topics/snyk-platform.md>)

Tags: [anthropic](<https://devfeed.tech/tags/anthropic.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-generative-ai](<https://devfeed.tech/tags/artificial-intelligence-generative-ai.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-fable-5-2](<https://devfeed.tech/tags/claude-fable-5-2.md>), [coding](<https://devfeed.tech/tags/coding.md>), [edition-us](<https://devfeed.tech/tags/edition-us.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [language-en](<https://devfeed.tech/tags/language-en.md>), [mythos-5-2](<https://devfeed.tech/tags/mythos-5-2.md>), [prices](<https://devfeed.tech/tags/prices.md>), [storytype-news](<https://devfeed.tech/tags/storytype-news.md>)

### AI overview

Anthropic says Fable 5.2 and Mythos 5.2 share a model but have different safeguards: Fable is broadly available, while direct Mythos access is limited to vetted cybersecurity and life sciences organizations. Both cost $8 per million input tokens and $40 per million output tokens, 20% below prior prices, and Anthropic says they generate output up to 30% faster. Fable 5.2 scored 49.1% on FrontierCode 2.0 at Xhigh effort, versus 32.4% for Fable 5.1; Anthropic also reported a comparable score to Opus 5.5 at roughly half the per-task cost at the specified effort settings. The article cautions that savings depend on workload and effort, and that agents still need monitoring, isolation, audit visibility, and human approval for critical changes. Mythos improved vulnerability rediscovery in Anthropic's internal evaluation, but could aid attackers as well as defenders. Eligible customers retain zero-data-retention terms; others face 30-day safety-monitoring retention. Enterprise Frontier Safeguards, in private beta, can monitor prompts and responses stored within customer infrastructure.

### Source excerpt

Anthropic has introduced Claude Fable 5.2 and Mythos 5.2, claiming significant improvements in handling complex software engineering tasks while cutting API prices by 20% compared with the previous versions. Fable 5.2 and Mythos 5.2 share the same underlying model but have different safeguards. Fable is broadly available, while direct access to Mythos is restricted to vetted cybersecurity and life sciences organizations to limit potential misuse. The models are both priced at $8 per million input tokens and $40 per million output tokens, down from $10 and $50, respectively. Anthropic said the newer versions can generate output up to 30% faster than their predecessors. The company reported substantial gains in coding performance. Fable 5.2 scored 49.1% on the FrontierCode 2.0 benchmark at its Xhigh effort setting, compared with 32.4% for Fable 5.1. The benchmark assesses whether maintainers of open-source projects would accept AI-generated changes to their code. Performance and costs could entice enterprises The models' improved performance and lower prices could make long-running AI agents viable for software migrations and scientific research, according to Lian Jye Su, chief analyst at Omdia. Higher task completion rates could further reduce deployment costs, he said. Anthropic said Fable 5.2 at its "High effort" setting achieved a coding benchmark score comparable to Opus 5.5 at its highest setting, at roughly half the cost per task. The comparison suggests that Fable could be more economical for some demanding assignments, although the savings would depend on the workload and effort setting. "Anthropic has cut the price of a token, not the price of a thought," said Sanchit Vir Gogia, chief analyst at Greyhound Research. He urged CIOs to treat AI processing effort as a budget consideration and measure costs against successfully completed tasks. Anthropic also recommends Opus 5.5 for most coding and knowledge work, with Fable and Mythos intended for more demanding

## An Agentic Multi-Agent Framework for System Log Anomaly Detection

DevFeed: [An Agentic Multi-Agent Framework for System Log Anomaly Detection](<https://devfeed.tech/articles/an-agentic-multi-agent-framework-for-system-log-anomaly-detection-126932.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.10612>)

Author: Ngoc-Son Le, Huy-Hoang Mac Nguyen, Ngoc Hong Tran

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Cryptography and Security](<https://devfeed.tech/sources/arxiv-cryptography-and-security.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Agentic SOC](<https://devfeed.tech/topics/agentic-soc.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-framework](<https://devfeed.tech/tags/ai-framework.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [anomaly-detection](<https://devfeed.tech/tags/anomaly-detection.md>), [attack](<https://devfeed.tech/tags/attack.md>), [cases](<https://devfeed.tech/tags/cases.md>), [challenge](<https://devfeed.tech/tags/challenge.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [cs-cr](<https://devfeed.tech/tags/cs-cr.md>), [detection](<https://devfeed.tech/tags/detection.md>), [diversity](<https://devfeed.tech/tags/diversity.md>), [effectively](<https://devfeed.tech/tags/effectively.md>), [efficiently](<https://devfeed.tech/tags/efficiently.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [framework](<https://devfeed.tech/tags/framework.md>), [log](<https://devfeed.tech/tags/log.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [system](<https://devfeed.tech/tags/system.md>)

### AI overview

The paper proposes a system log anomaly analysis workflow built with n8n and Gemini, organized into seven AI agent layers. Each agent handles a specific task and passes its result to the next layer, with the stated aim of speeding attack forensics and handling multiple log analysis requests concurrently. The authors report testing the workflow on 16 cases and evaluating per-case accuracy and operational performance; the abstract describes the findings as promising but gives no metrics or detailed results.

### Source excerpt

arXiv:2611.10612v1 Announce Type: new Abstract: As enterprise systems grow in scale and complexity, manual log analysis and conventional SIEM tools struggle to keep pace with the volume and diversity of incidents. To address this challenge, this paper proposed a multi-agent AI framework supporting in analyzing the system log to trace the attack forensic in the system more effectively and efficiently. The framework is designed using n8n platform and Gemini large language model. The framework consists of 7 AI agent layers. Each agent work on a specific task. When the task is done, the agent will forward the result to the output-associated agent of the next layer. The multi agent model improves the analysis velocity, so it is able to support numerous log analysis requirement at the same time. We deployed 16 test cases to test operation of the multi agent workflow, and evaluate the accuracy per test case and the operative performance. The result findings are very promising.

## WorkflowArena: Benchmarking Executable Workflow Orchestration in LLM-Based Multi-Agent Systems

DevFeed: [WorkflowArena: Benchmarking Executable Workflow Orchestration in LLM-Based Multi-Agent Systems](<https://devfeed.tech/articles/workflowarena-benchmarking-executable-workflow-orchestration-in-llm-based-multi-agent-systems-127276.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.12856>)

Author: Zheng Jiang, Fang Niu, Bo Peng, Houde Qian, Wenyi Li, Deyao Hong, Mingju Gao, Yizhe Chi, Youjie Zheng, Kaisen Yang, Calvin Xiao, Lifeng Sun, Huan-ang Gao

Published: 2026-11-10T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Software Engineering](<https://devfeed.tech/sources/arxiv-software-engineering.md>)

Topics: [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [claude-opus-5-5](<https://devfeed.tech/tags/claude-opus-5-5.md>), [cs-cl](<https://devfeed.tech/tags/cs-cl.md>), [cs-se](<https://devfeed.tech/tags/cs-se.md>), [delegation](<https://devfeed.tech/tags/delegation.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [legal-analysis](<https://devfeed.tech/tags/legal-analysis.md>), [mathematics](<https://devfeed.tech/tags/mathematics.md>), [multi-agent-systems](<https://devfeed.tech/tags/multi-agent-systems.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>), [workflow-orchestration](<https://devfeed.tech/tags/workflow-orchestration.md>)

### AI overview

WorkflowArena measures workflow orchestration as a distinct capability: whether an agent can decide when and how to delegate and organize work while producing an integrated deliverable. Its 145 long-horizon tasks span eight professional domains; hidden verifiers assess deliverable quality, and recorded delegation enables evaluation of work organization. Across nine models, no configuration led in every domain. At high effort, Claude models in Claude Code used workflows less often with larger worker groups, while GPT models in Codex delegated more often in smaller units. On a challenging subset with the executing model held fixed, workflows authored by Claude Opus 5.5 substantially improved weaker executors, suggesting that orchestration can affect task performance.

### Source excerpt

arXiv:2611.12856v1 Announce Type: cross Abstract: Large language model agents are increasingly deployed as multi-agent systems rather than as a standalone model operating within a single context. Recent agent harnesses allow the model to instantiate worker agents and to author the executable workflow that coordinates them, shifting the design of system structure from developers to the model at run time. The quality of this organization therefore constitutes a distinct capability, yet existing benchmarks either measure end-to-end task completion or evaluate coordination under structures specified in advance. We formalize this capability as workflow orchestration and introduce WorkflowArena, a benchmark of 145 long-horizon tasks spanning 8 professional domains, from software engineering and formal mathematics to legal analysis and 3D production. Each task specifies an initial workspace and an integrated deliverable composed of interdependent components, which is assessed by a hidden verifier. Agents determine autonomously whether and how to delegate, and all delegation is recorded, enabling joint evaluation of deliverable quality and the organization of the underlying work. Experiments with nine models show that no single configuration performs best across all domains and that systems exhibit distinct delegation strategies: at high effort, Claude models in Claude Code invoke workflows less frequently but with larger worker groups, whereas GPT models in Codex delegate more frequently in smaller units. When the executing model is held fixed on a challenging subset of tasks, workflows authored by Claude Opus 5.5 substantially improve weaker executors, indicating that workflow organization is itself a measurable determinant of task performance.

## EvoAgent: An Evolvable Agent Framework with Skill Learning and Multi-Agent Delegation

DevFeed: [EvoAgent: An Evolvable Agent Framework with Skill Learning and Multi-Agent Delegation](<https://devfeed.tech/articles/evoagent-an-evolvable-agent-framework-with-skill-learning-and-multi-agent-delegation-124393.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2604.20133>)

Author: Aimin Zhang, Jiajing Guo, Fuwei Jia, Chen Lv, Boyu Wang, Fangzheng Li

Published: 2026-11-09T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Multi-step AI processes](<https://devfeed.tech/topics/multi-step-ai-processes.md>)

Tags: [agent-architecture](<https://devfeed.tech/tags/agent-architecture.md>), [agent-delegation](<https://devfeed.tech/tags/agent-delegation.md>), [agent-framework](<https://devfeed.tech/tags/agent-framework.md>), [complex](<https://devfeed.tech/tags/complex.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [generation](<https://devfeed.tech/tags/generation.md>), [gpt](<https://devfeed.tech/tags/gpt.md>), [harness](<https://devfeed.tech/tags/harness.md>), [improvements](<https://devfeed.tech/tags/improvements.md>), [learning](<https://devfeed.tech/tags/learning.md>), [llm-as-judge](<https://devfeed.tech/tags/llm-as-judge.md>), [matching](<https://devfeed.tech/tags/matching.md>), [memory](<https://devfeed.tech/tags/memory.md>), [model](<https://devfeed.tech/tags/model.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [score](<https://devfeed.tech/tags/score.md>), [skills](<https://devfeed.tech/tags/skills.md>)

### AI overview

EvoAgent combines structured, multi-file skills with hierarchical sub-agent delegation to support task decomposition and ongoing capability improvement through user feedback. Skills include triggers and evolutionary metadata, and a three-stage matching strategy selects them; only persistent user-owned profile, memory, and skills are self-modified. In real-world foreign trade scenarios, integrating EvoAgent with GPT-5.2 raised the average score by about 28% under a five-dimensional LLM-as-Judge evaluation. Model-transfer experiments suggest performance also depends on how well the model and agent architecture work together. The evaluation corpus contains proprietary business data and is unavailable, so the reported results are for reference only.

### Source excerpt

arXiv:2604.20133v4 Announce Type: replace Abstract: This paper proposes EvoAgent--an evolvable large language model (LLM) agent framework that integrates structured skill learning with a hierarchical sub-agent delegation mechanism. EvoAgent models skills as multi-file structured capability units equipped with triggering mechanisms and evolutionary metadata, and is designed to enable continuous skill generation and optimization through a user-feedback-driven closed-loop process. In addition, by incorporating a three-stage skill matching strategy and organizing the agent as a harness whose only self-modified component is a persistent, user-owned state (profile, memory and skills), the framework supports dynamic task decomposition for complex problems and long-term capability accumulation. Experimental results based on real-world foreign trade scenarios demonstrate that, after integrating EvoAgent, GPT-5.2 achieves substantial improvements in professionalism, accuracy, and practical utility. Under a five-dimensional LLM-as-Judge evaluation protocol, the overall average score increases by approximately 28%. Further model transfer experiments indicate that the performance of an agent system depends not only on the intrinsic capabilities of the underlying model, but also on the degree of synergy between the model and the agent architecture. Code and documentation are released at https://github.com/Focus-AI-Center/Mentarc-EvoAgent; the evaluation corpus contains proprietary business data that cannot be released, so the reported results are for reference only.

## HarnessNet: Feedback-driven Hierarchical Joint Harness Evolution for Multi-agent Systems

DevFeed: [HarnessNet: Feedback-driven Hierarchical Joint Harness Evolution for Multi-agent Systems](<https://devfeed.tech/articles/harnessnet-feedback-driven-hierarchical-joint-harness-evolution-for-multi-agent-systems-124487.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.15221>)

Author: Zhaoyuan Qiu, Sachith Seneviratne, Yu Xia, Saman Halgamuge

Published: 2026-11-09T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [arxiv](<https://devfeed.tech/tags/arxiv.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [environment](<https://devfeed.tech/tags/environment.md>), [evolution](<https://devfeed.tech/tags/evolution.md>), [execution](<https://devfeed.tech/tags/execution.md>), [experiments](<https://devfeed.tech/tags/experiments.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [foundation-model](<https://devfeed.tech/tags/foundation-model.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [framework](<https://devfeed.tech/tags/framework.md>), [global](<https://devfeed.tech/tags/global.md>), [graph](<https://devfeed.tech/tags/graph.md>), [harness](<https://devfeed.tech/tags/harness.md>), [internals](<https://devfeed.tech/tags/internals.md>), [layer](<https://devfeed.tech/tags/layer.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [multi-agent-systems](<https://devfeed.tech/tags/multi-agent-systems.md>)

### AI overview

HarnessNet evolves multi-agent systems by dividing joint harness improvement across local optimizers. Each Agentron combines an evolvable harness, a frozen foundation model, and an optimizer that can inspect and modify only that harness. In a layered graph, environment feedback travels backward as node-specific signals, coordinating smaller optimization tasks without requiring each optimizer to access all agent trajectories or raw feedback. On AgentBench with multiple foundation models, the authors report performance gains comparable to a single global optimizer, with fewer tokens and less time per optimizer call while preserving local access boundaries. A migration study found the evolved harnesses remained jointly effective after foundation models were replaced, suggesting transfer across models.

### Source excerpt

arXiv:2610.15221v2 Announce Type: replace Abstract: AI agents can evolve by using environment feedback to revise their harnesses, the software managing model inputs, tool use, memory, and execution, while keeping foundation models frozen. Existing studies focus on single agents, leaving joint harness evolution in multi-agent systems (MAS) underexplored. A single global optimizer faces growing context and workload as the system scales, while access restrictions may prevent sharing agent internals or system-wide feedback. We propose HarnessNet, a framework for hierarchical harness evolution in MAS. Its basic unit, Agentron, comprises an evolvable forward harness, a foundation model, and a local optimizer that can observe and modify only that harness. Agentrons form a layered graph through which environment feedback propagates backward, with upstream optimizers receiving node-specific feedback from the downstream layer. This decomposes joint harness evolution into smaller, coordinated tasks with separate contexts, reducing individual optimizers' context and workload while keeping access local. Optimizers need not access all agent trajectories or raw environment feedback. Experiments on AgentBench with multiple foundation models demonstrate the feasibility of training HarnessNet. Compared with a single global optimizer, HarnessNet achieves comparable performance gains while respecting local privacy and requiring fewer tokens and less time per optimizer call. A migration study shows that the evolved harnesses remain jointly effective after replacing the foundation models, suggesting transferability across models.

## How to Build Multi-Agent Systems with the OpenAI Agents SDK

DevFeed: [How to Build Multi-Agent Systems with the OpenAI Agents SDK](<https://devfeed.tech/articles/how-to-build-multi-agent-systems-with-the-openai-agents-sdk-122659.md>)

Original publisher: [Read original article](<https://towardsdatascience.com/how-to-build-multi-agent-systems-with-the-openai-agents-sdk/>)

Author: Shuai Guo

Published: 2026-11-08T15:00:00Z

Content type: tutorial

Language: en

Sources: [Towards Data Science](<https://devfeed.tech/sources/towards-data-science.md>)

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [supervisor agent pattern](<https://devfeed.tech/topics/supervisor-agent-pattern.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [agents-sdk](<https://devfeed.tech/tags/agents-sdk.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [blocks](<https://devfeed.tech/tags/blocks.md>), [build](<https://devfeed.tech/tags/build.md>), [building](<https://devfeed.tech/tags/building.md>), [data](<https://devfeed.tech/tags/data.md>), [editor-s-picks](<https://devfeed.tech/tags/editor-s-picks.md>), [fan](<https://devfeed.tech/tags/fan.md>), [first](<https://devfeed.tech/tags/first.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [multi-agent](<https://devfeed.tech/tags/multi-agent.md>), [multi-agent-systems](<https://devfeed.tech/tags/multi-agent-systems.md>), [openai](<https://devfeed.tech/tags/openai.md>), [openai-agents](<https://devfeed.tech/tags/openai-agents.md>), [openai-agents-sdk](<https://devfeed.tech/tags/openai-agents-sdk.md>), [post](<https://devfeed.tech/tags/post.md>), [python](<https://devfeed.tech/tags/python.md>), [sdk](<https://devfeed.tech/tags/sdk.md>), [support](<https://devfeed.tech/tags/support.md>), [systems](<https://devfeed.tech/tags/systems.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

The excerpt distinguishes two OpenAI Agents SDK delegation patterns. With `specialist_agent.as_tool(...)`, a coordinating agent calls specialists for bounded subtasks, stays active, and combines their results; the EV purchase adviser uses vehicle, charging, and ownership-cost researchers this way. With `handoffs=[...]`, the initial agent routes a request and the specialist takes control and produces the final response; the EV support agent routes a public-charging question to a charging specialist. The excerpt presents fan-out/fan-in as centralized parallel work, pipelines as sequential chains, hierarchies as recursive agent-as-tool relationships, evaluator-optimizers as centralized loops, and swarms as dynamic peer-to-peer handoffs. These patterns combine delegation, topology, and execution choices.

### Source excerpt

Build an EV purchase adviser and a support agent in Python, then see how fan-out, pipelines, hierarchies, and swarms reuse the same building blocks. The post How to Build Multi-Agent Systems with the OpenAI Agents SDK appeared first on Towards Data Science.

[Next page](<https://devfeed.tech/topics/hierarchical-agent-systems.md?cursor=WyIyMDI2LTExLTA4VDE1OjAwOjAwKzAwOjAwIiwgIjRlNGFmYjM5LThmOGMtNDhhZi04Y2MyLWNjYWM5OTdjNjk0MCJd>)