[DOCS] Auto memory docs omit the error when `MEMORY.md` exceeds its read limit

Documentation Type

Missing documentation (feature not documented)

Documentation Location

https://code.claude.com/docs/en/memory

Section/Topic

The Auto memory section, especially How it works and the guidance for keeping MEMORY.md concise.

Current Documentation

The auto-memory documentation currently says:

MEMORY.md acts as an index of the memory directory. Claude reads and writes files in this directory throughout your session, using MEMORY.md to keep track of what's stored where.

It also says:

The first 200 lines of MEMORY.md, or the first 25KB, whichever comes first, are loaded at the start of every conversation. Content beyond that threshold is not loaded at session start. Claude keeps MEMORY.md concise by moving detailed notes into separate topic files.

What's Wrong or Missing?

The page documents the startup read limit and recommends moving detail into topic files, but it does not explain what happens when Claude writes to MEMORY.md and that write would leave the index over the limit.

Claude Code v2.1.210 now produces an explicit error for such memory writes instead of silently truncating the index. This is a user-visible change to the behavior of the documented automatic write process. Without it, users cannot tell whether an oversized write was saved, truncated, or rejected, nor how to recover when memory maintenance fails.

The persistent subagent-memory guidance has the same gap: it tells subagents to curate MEMORY.md when it exceeds the limit and enables the Write and Edit tools, but does not describe the explicit error introduced in v2.1.210.

Suggested Improvement

Add a versioned note to the How it works subsection explaining that, as of v2.1.210, a write that would leave a MEMORY.md index over the 200-line or 25KB read limit fails with an explicit error instead of silently truncating the file. Explain that the remedy is to keep the index concise by moving detailed notes into topic files before retrying the write.

Add the same behavior, or a cross-reference to the auto-memory explanation, to the persistent subagent-memory section so users configuring memory: user, project, or local understand the write limit there as well.

Impact

Medium - Makes feature difficult to understand

Additional Context

The v2.1.210 release notes state:

Memory writes that leave a MEMORY.md index over its read limit now produce an explicit error instead of silent truncation

Affected Pages:

| Page | Context |
|------|---------|
| https://code.claude.com/docs/en/memory | Auto memory storage and the MEMORY.md read limit |
| https://code.claude.com/docs/en/sub-agents | Persistent subagent memory and MEMORY.md curation |
| https://code.claude.com/docs/en/changelog | v2.1.210 release note describing the changed write behavior |

Total scope: 2 feature-documentation pages affected, with the changelog as the behavior reference.