mlsanityData to 14 Oct 2026

CodeNib: A Multi-View Data System for Serving Repository Context to Coding Agents

arXiv 2607.25431 ↗First posted 28 Jul 2026cs.SEcited by 0

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Coding agents repeatedly search, navigate, and retain context from evolving repositories, but disconnected indexes, language servers, and task-local histories force repeated discovery and obscure lifecycle costs. CodeNib builds reusable lexical, dense, and structural views per repository commit, maps outputs to repository-relative source ranges, maintains selected views across edits, and serves ranked search, symbol navigation, and bounded context through one runtime. Across 100 snapshots, we map quality-cost frontiers across the repository-context lifecycle. When outputs match an independent rebuild, graph and vector updates are 8.7× and 25.4× faster at the median. On the static-navigation subset matching normalized live-server locations (63% of 1,000 requests), the median per-request live/static latency ratio is 4.7×. Across five models, selected context policies preserve localization with 50--87% fewer trajectory tokens than paired grep/read. Together, these results support multi-view repository-context serving with explicit, operation-specific validity boundaries.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 39 of its 78 references that are in the map. First posted on arXiv 28 Jul 2026.

Builds on

39 of 78 references.

Closest ancestors · 13

Same direction, newest first

  1. 2026Code Isn't Memory: A Structural Codebase Index Inside a Coding Agent21 Jun
  2. 2026What Context Does a Coding Agent Actually Need to Act?19 Jun
  3. 2026SWE-Explore: Benchmarking How Coding Agents Explore Repositories5 Jun
  4. 2026CODESTRUCT: Code Agents over Structured Action Spaces7 Apr
  5. 2026ContextBench: A Benchmark for Context Retrieval in Coding Agents5 Feb
  6. 2026Repository Intelligence Graph: Deterministic Architectural Map for LLM Code Assistants15 Jan
  7. 2025Context as a Tool: Context Management for Long-Horizon SWE-Agents26 Dec
  8. 2025SpIDER: Spatially Informed Dense Embedding Retrieval for Software Issue Localization18 Dec
  9. 2025SpecAgent: A Speculative Retrieval and Forecasting Agent for Code Completion20 Oct
  10. 2025Reducing Cost of LLM Agents with Trajectory Reduction28 Sep
  11. 2025LocAgent: Graph-Guided LLM Agents for Code Localization12 Mar
  12. 2025OrcaLoca: An LLM Agent Framework for Software Issue Localization1 Feb
1 more
  1. 2024SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering6 May

Other references · 26

  1. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  2. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  3. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,373
  4. 2017Billion-scale similarity search with GPUscited by 2,497
  5. 2020CodeBERT: A Pre-Trained Model for Programming and Natural Languagescited by 1,310
  6. 2016Efficient and robust approximate nearest neighbor search using Hierarchical Navigable Small World graphscited by 861
  7. 2020GraphCodeBERT: Pre-training Code Representations with Data Flowcited by 609
  8. 2023RepoCoder: Repository-Level Code Completion Through Iterative Retrieval and Generationcited by 367
  9. 2022UniXcoder: Unified Cross-Modal Pre-training for Code Representationcited by 361
  10. 2026Gemma 4 Technical Reportcited by 259
  11. 2024RepoGraph: Enhancing AI Software Engineering with Repository-level Code Graphcited by 91
  12. 2024CodexGraph: Bridging Large Language Models and Code Repositories via Code Graph Databasescited by 65
14 more
  1. 2024Repoformer: Selective Retrieval for Repository-Level Code Completioncited by 55
  2. 2025The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Managementcited by 48
  3. 2026Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interactioncited by 38
  4. 2025Issue Localization via LLM-Driven Iterative Code Graph Searchingcited by 22
  5. 2025SweRank: Software Issue Localization with Code Rankingcited by 20
  6. 2025Efficient Code Embeddings from Code Generation Modelscited by 13
  7. 2025CoRet: Improved Retriever for Code Editingcited by 3
  8. 2026RepoShapley: Shapley-Enhanced Context Filtering for Repository-Level Code Completioncited by 3
  9. 2022Stack graphs: Name resolution at scalecited by 2
  10. 2026CodeMEM: AST-Guided Adaptive Memory for Repository-Level Iterative Code Generationcited by 2
  11. 2026On the Effectiveness of Context Compression for Repository-Level Tasks: An Empirical Investigationcited by 2
  12. 2026TypeScript Repository Indexing for Code Agent Retrievalcited by 2
  13. 2026AOCI: Symbolic-Semantic Indexing for Practical Repository-Scale Code Understanding with LLMscited by 2
  14. 2025LSPRAG: LSP-Guided RAG for Language-Agnostic Real-Time Unit Test Generationcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026FastContext: Training Efficient Repository Explorer for Coding Agents
  2. 2026RepoAtlas: Guiding Coding Agents via Evolving Multimodal Repository Views
  3. 2026Cost-Effective Repository Exploration for Agentic Issue Localization
  4. 2026LARGER: Lexically Anchored Repository Graph Exploration and Retrieval
  5. 2026ReCUBE: Evaluating Repository-Level Context Utilization in Code Generation
  6. 2025Prometheus: Towards Long-Horizon Codebase Navigation for Repository-Level Problem Solving
  7. 2026Agent Retrieval Bench: Evaluating Repository Context Retrieval for Coding Agents
  8. 2026Learning to Commit: Generating Organic Pull Requests via Online Repository Memory