Follow Discord

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

← All of v2.1.251 Home All releases ←olderv2.1.250

Claude Code v2.1.251 28 Aug 2026

Eval sandboxes run with claude.ai MCP servers off

Under the hood

Useful1 Signal3

Plugin Eval

Eval sandboxes force claude.ai MCP servers off so they can't reach real servers.

ENABLE_CLAUDEAI_MCP_SERVERS

What

The internal evaluation harness now forces ENABLE_CLAUDEAI_MCP_SERVERS to false and pins its own temp directories, so evals cannot reach real MCP servers or share scratch space with the host.

Details

Pins TMPDIR, TMP, TEMP and CLAUDE_CODE_TMPDIR.
Affects the eval harness only; it is not a runtime flag for normal use.

Evidence

ENABLE_CLAUDEAI_MCP_SERVERS: "false"

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

See this entry in the whole of v2.1.251 →