Reasoning effort, not tool access, buys first-try reliability in agentic code generation: an observational study
Agentic coding assistants are increasingly given extra capabilities, such as browser based testing tools and design oriented system prompts, on the assumption that more capability yields better software. This study tested that assumption directly. Ninety independent agent runs built the same application, a real time retrospective board, from one detailed specification, each scored on a fixed 14 criterion functional rubric (42 point maximum) and a visual quality review. The runs spanned several model generations, two agent harnesses, two reasoning effort levels, a testing tool, and two design o
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 62%AgentTrace →
- PossiblePossibly related (embedding) · 60%AgentCore-8B →
- PossiblePossibly related (embedding) · 56%digiteinfotech/kairon →
- PossiblePossibly related (embedding) · 56%Open-source agent framework crosses 50k stars →
- PossiblePossibly related (embedding) · 56%patrick-toulme/harnessgym →
- LinkedLinked via arxiv author · 85%Achint Mehta →
“Reasoning effort, not tool access, buys first-try reliability in agentic code generation: an observational study”
- PossiblePossibly related (embedding) · 48%anatomia-dev/anatomia →
- PossiblePossibly related (embedding) · 47%go-appsec/toolbox →