# Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding

arXiv: https://arxiv.org/abs/2608.11095 · HTML: https://mlsanity.com/paper/2608.11095/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Agent memory systems, one of 16 directions in LLM agents and tool use.
- Builds on 15 of its 43 references that are in the map.
- First posted on arXiv 11 Aug 2026.

## Placed in

- LLM agents and tool use / [Agent memory systems](https://mlsanity.com/agents/agent-memory-systems/)

## Builds on (most cited first, 15 of 15)

- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [MemGPT: Towards LLMs as Operating Systems](https://mlsanity.com/paper/2310.08560/) (2023)
- [Learning to Prompt for Continual Learning](https://mlsanity.com/paper/2112.08654/) (2021)
- [Large Language Models Can Be Easily Distracted by Irrelevant Context](https://mlsanity.com/paper/2302.00093/) (2023)
- [DualPrompt: Complementary Prompting for Rehearsal-free Continual Learning](https://mlsanity.com/paper/2204.04799/) (2022)
- [CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learning](https://mlsanity.com/paper/2211.13218/) (2022)
- [Zep: A Temporal Knowledge Graph Architecture for Agent Memory](https://mlsanity.com/paper/2501.13956/) (2025)
- [Learning to Update Natural Language Comments Based on Code Changes](https://mlsanity.com/paper/2004.12169/) (2020)
- [Agent READMEs: An Empirical Study of Context Files for Agentic Coding](https://mlsanity.com/paper/2511.12884/) (2025)
- [Deep Just-In-Time Inconsistency Detection Between Comments and Source Code](https://mlsanity.com/paper/2010.01625/) (2020)
- [What Makes a Good Commit Message?](https://mlsanity.com/paper/2202.02974/) (2022)
- [Prompting in the Wild: An Empirical Study of Prompt Evolution in Software Repositories](https://mlsanity.com/paper/2412.17298/) (2024)
- [On the Impact of AGENTS.md Files on the Efficiency of AI Coding Agents](https://mlsanity.com/paper/2601.20404/) (2026)
- [WildIFEval: Instruction Following in the Wild](https://mlsanity.com/paper/2503.06573/) (2025)
- [FSFM: A Biologically-Inspired Framework for Selective Forgetting of Agent Memory](https://mlsanity.com/paper/2604.20300/) (2026)

## Built on by (most cited first, 3 of 3)

- [From Agent Behaviour to Agent-Friendly Documentation: An Empirical Study of How Coding Agents Discover, Read, and Write Technical Documentation](https://mlsanity.com/paper/2608.20195/) (2026)
- [Curating Always-Loaded Context for LLM Agents: A Capacitated Assortment Model with Censored Feedback](https://mlsanity.com/paper/2610.11007/) (2026)
- [Cortex: Measurement-Gated Evolution and Pruning of Coding-Agent Skills and Rules](https://mlsanity.com/paper/2610.14216/) (2026)

## Abstract (arXiv)

Agentic coding READMEs like CLAUDE.md grow without bound in real repositories, stopping only when the repository retires or someone rewrites the file wholesale. We trace this to imperfect recall: appending an instruction is always cheap, but once an instruction's rationale is gone, deleting it without risking a correctness regression costs O(2^|D|) in a prompt of |D| instructions. We name the resulting divergence catastrophic remembering, the inverse of catastrophic forgetting around which continual learning is organized. First, we characterize this phenomenon across 247,694 instruction lifetimes in 1,867 repositories: agentic prompts grow without bound, more than tripling over their lifetime (+226%), gaining +4.9 net instructions every commit; further, the older an instruction gets, the less likely it is to be deleted (log-hazard -0.032/commit). Then, we show that prompt comments can halt the growth: inverting IFEval yields verifiable worlds whose optimal prompts are known, and there comments encoding latent reasoning remove 99.3% of excess instructions (+211.3% to +1.4%). Finally, applying the same inversion to WildIFEval, we show that prompt comments can improve real-world agentic instruction-following by up to 23.1%. If English is the new code, why don't we have comments yet?