Claude Code v2.1.225~v2.1.226 Major Updates - SendMessage Cross-Machine Origin Support and Authentication-Related Fixes

This is Ishikawa from the Cloud Business Division. I'll be summarizing the Claude Code updates for v2.1.225 ~ v2.1.226 (2026-08-07 ~ 2026-08-08). There aren't many new features, but there are fixes related to authentication and session handling, with no breaking changes. Today I tried out "SendMessage initiated from another machine."

The previous update article is here.

https://dev.classmethod.jp/articles/20260807-cc-updates-v2-1-224/

Update Summary

This covers 2 versions (v2.1.225 ~ v2.1.226, 2026-08-07 ~ 2026-08-08), with 15 changes listed in the CHANGELOG. The breakdown is 2 new features, 4 security/authentication-related items, 1 improvement, and 8 fixes, with no breaking changes or deprecations. The majority of changes are concentrated in v2.1.225, centered around stabilization of Remote Control and authentication.

Notable Updates

SendMessage Can Now Initiate to Sessions on Other Machines (v2.1.225)

Continuing from the recent release (v2.1.225), SendMessage can now initiate conversations by specifying the name of a Remote Control session running on another machine. In ListAgents, these are displayed as name [ref]. Previously, you could only reply after receiving a message from the other party.

Since you can now reach out first to a session on another machine, those who divide work across multiple machines will find this useful.

Usage Warnings Now Support Gateway Spend Limits (v2.1.225)

Usage warnings now support gateway spend limits, and when a limit is reached, the message now displays the limit value, reset time, and a message from the operator. The gateway also needs to be on v2.1.225.

Since the display when a limit is reached now shows the limit value, reset time, and operator message side by side, those who operate a gateway internally should find that support inquiries are reduced.

Workspace Trust Prompt Added to claude agents (v2.1.225)

When launching claude agents for an untrusted directory, a workspace trust confirmation prompt is now displayed, bringing it in line with the behavior of claude.

For those who frequently run background agents in directories of other projects, having the same confirmation as claude is meaningful.

Fixed Bug Where CLAUDE_CODE_OAUTH_TOKEN Was Replaced with Short-Lived Token (v2.1.225)

A bug was fixed where a temporary 401 error caused a long-lived CLAUDE_CODE_OAUTH_TOKEN to be replaced by a short-lived token from saved login credentials, breaking headless sessions until restart.

For those using CLAUDE_CODE_OAUTH_TOKEN in CI or batch processing, since this was the kind of failure that wouldn't fix itself until a restart, it's worth prioritizing this upgrade.

Fixed Bug Where MCP OAuth Server Repeatedly Returns 401 on macOS (v2.1.225)

A bug was fixed where, after a keychain read timeout on macOS, the MCP OAuth server would repeatedly return 401 errors as if it had never been authenticated, causing intermittent failures.

Those who regularly use OAuth MCP servers on macOS and have occasionally seen authentication appear to drop out may find this fix explains what was happening.

Fixed Bug Where Conversation History Gets Corrupted on Remote Control Session Resume (v2.1.225)

A bug was fixed where resuming a Remote Control session after a very large conversation had been compacted would corrupt the conversation history.

For those who use Remote Control while continuing long conversations, being able to avoid discovering history corruption only after resuming is a significant improvement.

Target Versions and Period

New Features

• You can now initiate a conversation from SendMessage to a Remote Control session on another machine by specifying its name (v2.1.225)
• Usage warnings now support gateway spend limits, displaying the limit value, reset time, and operator message (gateway also needs to be on v2.1.225) (v2.1.225)

Security / Authentication

• A workspace trust confirmation prompt for untrusted directories has been added to claude agents (v2.1.225)
• Fixed a bug where a temporary 401 would cause a long-lived CLAUDE_CODE_OAUTH_TOKEN to be replaced by a short-lived token from saved login credentials (v2.1.225)
• Fixed a bug on macOS where the MCP OAuth server would repeatedly return 401 errors after a keychain read timeout (v2.1.225)
• Fixed a bug in auto mode where a self-permission check rejected by the safety filter was being counted toward the consecutive block count. The operation itself continues to be rejected, but the model is now told to move on rather than retry (v2.1.225)

Improvements

• In Remote Control, photos attached from the Claude app are now passed directly to Claude instead of being read separately from disk via a tool call (v2.1.225)

Fixes (Key Items)

Highlights from fixes related to stability and usability.

• Fixed bug where conversation history gets corrupted on Remote Control resume: The issue where conversation history was corrupted when resuming a session after a very large conversation had been compacted has been resolved (v2.1.225)
• Fixed stalling of inter-session messages: The issue where inter-session messages were left pending indefinitely without notification or expiration, in headless sessions and at startup, has been resolved (v2.1.225)
• Fixed handling of self-hosted-runner startup failures: When claude self-hosted-runner cannot create or write to --base-dir, instead of completing registration while all sessions continue to fail, it now exits with a clear error at startup (v2.1.225)
• Fixed false session detection for Claude Code on the web: The issue where sessions were incorrectly reported as stopped and a growing backlog of events was resent on every reconnect has been resolved (v2.1.225)
• Fixed collapsing in VSCode Focus view: The issue where the latest TODO list, pending question context, and confirmed answers were being collapsed has been resolved; thinking-only collapses now display as "Thought for Ns" and collapse again when that turn completes (v2.1.225)
• Fixed wrong destination in SendMessage: Once a Remote Control destination is confirmed, it will no longer be replaced by a local session with the same name even if the list cannot be re-checked (v2.1.225)
• A quietly welcome fix: Fixed a bug where simply hovering over a session from another project in the agent list would change the starting directory for the next agent to launch (v2.1.225). Since changing the working location of the next agent just by hovering is the kind of accident that's hard to notice, this feels like a fix that works quietly in the background.
• Additionally, v2.1.226 includes bug fixes and reliability improvements with no individual line items listed.

Trying Out "SendMessage Initiated from Another Machine"

I'll verify the point where ListAgents in v2.1.225 now displays sessions in name [ref] format, within a read-only scope.

I prepared by running claude in a separate terminal in advance and used /rename session-a to rename the session for easier identification.

Since actually sending via SendMessage would affect the other session, I won't execute that and will only look at the list display.

# Run under a temporary directory (only ListAgents in read-only mode is permitted)
% echo "Please run the ListAgents tool." | claude -p --max-turns 1 \
  --allowedTools ListAgents --output-format stream-json --verbose > listagents.jsonl

# Extract only the tool execution results from the output JSON Lines
% jq -r 'select(.type=="user") | .message.content[]? | select(.type=="tool_result") | .content' listagents.jsonl
Peer sessions (1):
  session-a [845f99]  ·  interactive  ·  idle  ·  started 25s ago

You can confirm that a reference like [845f99] is appended after the session name, and sessions are listed in name [ref] format. Note that claude's exit code will be 1, but this is because --max-turns 1 was reached; the ListAgents execution itself succeeded.

Previously you had to wait for the other party to speak first, but being able to assign work by specifying a name proactively feels like it increases flexibility when working across multiple machines. Since I didn't have Remote Control connected locally, I wasn't able to confirm the rows for other machines, but the display format was consistently name [ref].

Closing

These 2 versions feel more focused on solidifying the foundation around authentication and session handling than on new features. The more you rely on headless execution or MCP, the more value there is in upgrading.

If there are any changes that interest you, try upgrading and checking them out.

References

https://github.com/anthropics/claude-code/blob/main/CHANGELOG.md

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

https://dev.classmethod.jp/articles/20260807-cc-updates-v2-1-224/