Summary
Choosing the right design patterns for software agents can be tricky because engineers need more than just options—they need evidence to know what works. The authors combine real-world cases with controlled experiments to test a new way of selecting and combining these patterns using contracts and measurable results. Their method succeeded much more often in completing tasks or safely refusing them than simpler baseline methods. However, some tricky cases still cause failures, especially when assumptions about trustworthy evidence fail. This study shows practical benefits of structured pattern use but does not claim to outperform all existing decision methods.
What this means in practice
- •For software architects: Use the evidence-driven selection procedure to improve reliability when designing multi-agent systems with complex pattern compositions.
- •For industrial control teams: Apply bounded composition mechanisms to safely manage failure and recovery scenarios in automated system controls requiring agent collaboration.
Authors
Jia Huang, Bingsheng Ru, Bo Liang, Yuke Xiong, Haiyan Fan, Joey Tianyi Zhou
Abstract
Agent pattern catalogs describe architectural alternatives, but engineers also need evidence for selecting and composing them. We extend a cognitive-function--execution-topology framework with an evidence-driven selection procedure and explicit composition contracts. The study combines two industrial case accounts, a contributor's collaboration design, seven practitioner workshops, and executable fault-injection experiments. Three local reference systems use SQLite state, shared artifact files, and a file-backed map-publication surrogate. A 2 x 2 ablation separates function-specific contracts from execution controls across 36 scenario families, with ten input variants per family. Under equal family weighting, composition correctly completes or safely refuses 11 of 12 families in each system, compared with 2, 2, and 1 for the simple baselines. These are constructed challenge-suite results, not production success rates. Recovery without consumer-side verification increases false-success cases in the map surrogate. All composed systems still fail a boundary case in which trusted evidence is wrong. An audit of a frozen six-arm language-model pilot reproduces its scores but confirms an evaluator-dependent ranking reversal. The contribution connects industrial failure, pattern selection, executable treatment, external acceptance, and revision. Selection is demonstrated rather than evaluated against an alternative procedure; catalog admission rests on separate criteria. The results support bounded composition mechanisms, not a claim that the framework improves engineers' design decisions over an equally informative conventional method.