Large-scale Repository Engineering via Agent-Native Reusable Code Primitives

Key points

• Large language models equipped with development environments have moved code generation toward repository-scale construction, yet building complete repositories remains difficult because interacting modules, interfaces, configurations, tests, and dependencies must work together.1
• We introduce Code Primitives, agent-native reusable executable components with interface contracts, dependency closures, validation tests, and provenance.1
• Each primitive uses a resident LLM to assess relevance and adapt its implementation, interfaces, and dependencies to the target repository, and we organize 1,424 validated primitives in CodeFace, a searchable library for repository construction.1
• To measure construction end to end, we build LEGO-REPO, a benchmark of 522 executable reconstruction tasks spanning seven software domains, 22 capability tracks, and five difficulty levels, scored against native test suites between an empty-package floor and original-source ceiling.1

Sources (1)

• [1]Large-scale Repository Engineering via Agent-Native Reusable Code Primitives
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 08:20 PM

  We introduce LEGO (Large-scale repository Engineering via aGent-native reusable cOde primitives), which activates task-relevant primitives, integrates their adapted implementations with task-specific code while resolving cross-component constraints, and revises the result against executed tests.

  Large language models equipped with development environments have moved code generation toward repository-scale construction, yet building complete repositories remains difficult because interacting modules, interfaces, configurations, tests, and dependencies must work together.

Extractive summary: sentences quoted from the sources.