When Instruction Files Go Stale: Documentation Debt and Hidden Cost in Coding Agents
cs.SE, cs.AI
Submitted: 2026-11-06
Updated: 2026-11-06
Code: https://github.com/ZJU-PL/pyflow
Terminology
Sources
- Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?
- On the Impact of AGENTS.md Files on the Efficiency of AI Coding Agents
- Context Engineering for AI Agents in Open-Source Software
- Agent READMEs: An Empirical Study of Context Files for Agentic Coding
- Context Rot in AI-Assisted Software Development: Repurposing Documentation Consistency for AI Configuration Artifacts
- Stale Instruction Files Override the Code in Coding Agents: A One-Sentence Fix
- Harness Engineering for Agentic AI Coding Tools: An Exploratory Study
- Probe-and-Refine Tuning of Repository Guidance for Coding Agents
- SWE-ABS: Adversarial Benchmark Strengthening Exposes Inflated Success Rates on Test-based Benchmark
- Probe to Generate: Program Variant-Guided Test Augmentation for Repository-Level Repair Benchmarks
- Understanding Software Engineering Agents: A Study of Thought-Action-Result Trajectories
- Beyond Resolution Rates: Behavioral Drivers of Coding Agent Success and Failure
- AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation
- What Resolve Rate Hides: Trajectory Structure Diagnostics for Coding Agents
- CodeUpdateArena: Benchmarking Knowledge Editing on API Updates
- Aletheia: Permission-Minimality Testing for Coding-Agent Rules
Related papers
- ArkEval: Benchmarking and Evaluating Automated CodeRepair for ArkTS
- TestPrism: Rethinking Test Evaluation Beyond a Single Reference
- Mining Workflow Graphs for Black-Box Boundary Testing of Conversational LLM Agents
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- For Coding Agents, the Repository Matters