AIToday
Large Language ModelsAI Coding AssistantsZenn AI/MLPublished: Oct 1, 2026, 22:00 JST

Claude Code 2.1.278: @AGENTS.md import works, stand-alone AGENTS.md fails

Claude Code 2.1.278: @AGENTS.md import works, stand-alone AGENTS.md fails

3 Key Points

  1. What happened

    The author tested Claude Code 2.1.170 and 2.1.278 in 12 tool-disabled runs on Windows; @AGENTS.md import returned both markers on both, but AGENTS.md alone returned NONE on 2.1.278.

  2. Why it matters

    The import path appears to be the one route that produced a marker across both versions, so the setup under test — not the feature itself — may be what decided the result.

  3. What to watch

    The lone-placement failure remains unexplained, since no internal load trace was captured and mode, tool settings and flag effects were not isolated. Watch the 2.1.277 native-support threshold.

WHO IT HITSDevelopers running Claude Code on their own machines and people who maintain shared instruction files in coding projects are the ones who would need to re-test their exact launch settings rather than assume version alone settles behavior.

Not sure about something? Ask the AI

Questions and answers are published on this page.

Summaries like this, in your inbox every morning.

Context & Analysis

The test exists because the author wanted to separate the official specification from what actually happens on one machine. The current documentation places native support at 2.1.277 and later, with conditions tied to the presence of other instruction files, plus feature enablement and first-session constraints. Rather than assume an updated build behaves the same everywhere, he fixed the versions and compared unique marker strings in responses.

The 12 runs span six conditions each on 2.1.170 and 2.1.278, and every run logged an empty tool, MCP and skills set with no actual tool calls. That matters because it closes off the ordinary route where the model simply opens a file and later repeats what it read. Under those conditions the @AGENTS.md import produced both markers on both builds, while the explicit both-files mode produced a marker only on the new build.

The unresolved point is the lone-placement case on 2.1.278, which returned NONE even though the same build surfaced the AGENTS.md marker when both files were read. The author is careful not to call this a verdict on native support; he simply did not capture the internal load path, and interactive mode, normal tool-enabled runs, user settings and the effect of flags such as --disable-slash-commands were left untested. For teams relying on shared instruction files, the practical reading is that behavior may hinge on the exact launch configuration rather than the version number alone, and a small marker-string check before changing a live project looks like the safer path.

FAQ
Did the test prove Claude Code 2.1.278 does not support AGENTS.md?
No. The author says the lone-placement NONE does not disprove native support, because the 2.1.278 run with both files read explicitly did return the AGENTS.md marker.
How was the test set up?
Each condition used an empty directory with only a unique LOADMARK string in the files, prompts asked for that value, and tools were disabled with --tools "" so the model could not read files itself.
What does the author say the record actually shows?
Only what appeared in responses under a fixed launch setup. Each condition ran once, so follow-through rates, real task quality and cross-OS compatibility were not measured.

AI news that matters for your work, delivered every morning.

Pick your industry and the AI tools you use, and get news related to your work every day.

Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →

Ask AI

Ask AI anything about this article. The AI reads this article, earlier AIToday articles, and Wikipedia, and cites its sources. Q&As are published on this page for other readers too.

Questions and answers are published on this page.

Related Articles

Next articleOllaya runs decision models locally, up to 255 options