Skip to content
Agentsmedium signalverified

Claude Code 2.1.285 lets admins restrict API providers and fixes a token login that skipped org policy

Anthropic released Claude Code 2.1.285 on September 29, 2026. A new managed setting, allowedProviders, limits which API providers a machine may use. Two fixes matter for teams: sessions that authenticate with an auth token never loaded the organization's policy, and redacted logs could show a URL password. The release also adds a switch that turns off WebFetch and a command that opens the desktop app.

By Redakcija WebAiRadarPublished 2 min readwritten by a model

Source

Release v2.1.285

Claude Code Releases (GitHub) · Original published September 29, 2026

Claude Code 2.1.285 is mostly a fixes release, but the fixes cluster around one theme: what an organization's managed settings actually control. The release adds a provider allowlist, closes a gap where an auth-token login ran without organization policy, and changes what happens when the managed settings file cannot be read.

What managed settings now cover

The new allowedProviders managed setting limits which API providers a machine may use. The list in the notes is the Anthropic API, a custom endpoint, Bedrock, Mantle, Vertex AI, Foundry, Claude Platform on AWS, and a Cloud gateway. Before this, a managed machine could be pointed at any of them.

A second change is easy to misread. When the operating system denies reading the managed settings file, Claude Code used to refuse to start. From 2.1.285 it warns and starts without that file's policies. Other read errors, and a file that cannot be parsed, still stop every session. An administrator who relied on the hard failure now needs to watch for the warning instead.

Fixes that touch security

Sessions that authenticate against the Anthropic API with ANTHROPIC_AUTH_TOKEN never loaded the organization's policy. The notes describe this as fixed, without saying since when it applied. Three other fixes point the same way:

  • redacted logs and transcripts showed part of a URL password that contains @, or all of it when the URL writes @ as %40;
  • in SDK and -p sessions, switching off an MCP server added mid-session left its tools available;
  • fork subagents did not keep the session's plan mode or dontAsk mode, and a fork now runs under its parent's permission mode and cannot exit plan mode.

New switches and commands

The CLAUDE_CODE_DISABLE_WEB_FETCH environment variable turns off the WebFetch tool. claude --desktop opens the Claude desktop app on the current directory, or on a session with --continue or --resume.

Plugins get two additions. claude plugin configure <plugin> shows a plugin's options and which are unset, and claude plugin install --config accepts values in the form <server>.<key>=<value>, so a bundled .mcpb MCP server can be configured at install time instead of through the /plugin menu. Installing a plugin whose id differs from an installed one only in ., -, @, or letter case used to write into the other plugin's cache or data folder; the install is now refused.

One fix removes a daily annoyance: sandbox auto-allow asked for approval on every run of inline scripts such as python3 -c or node -e simply because they contained =.

„it now warns and starts without that file's policies“
Claude Code 2.1.285 release notes
BrandsClaude

Related

OPENCLAW 2026.10.232,768characters, and past that a key could slip thr
Agentsmedium signal

OpenClaw 2026.10.2 stops long tool output from slipping credentials past redaction

A bug in OpenClaw's secret masking let a credential through whenever it straddled the boundary between two slices of a long text. The fix ships in 2026.10.2 along with repaired updates from 2026.10.1 and a new compaction default for OpenAI and Anthropic API keys.

OpenClawverified

The OpenClaw web interface with a conversation on the left and the Bits & Bolts reference MCP app open beside it, showing a 3D keycap model and an Attach view + note button.
Agentsmedium signal

OpenClaw 2026.10.1 lets agents search installed skills and changes how sessions message each other

OpenClaw released version 2026.10.1 on October 10, 2026, its first stable release since 2026.9.7. Agents can now search the skills installed on a machine and read only the one a task needs, instead of carrying the whole library in the prompt. Opt-in MCP Apps run beside chat, and full backups can go to an external drive or Cloudflare R2. Three changes can break existing setups: session-to-session messaging, the old required argument, and Workboard data stored before July 2026.

OpenClawverified

CLAUDE CODE2.1.294hooks written as instructions now block what t
Agentsmedium signal

Claude Code 2.1.294 fixes prompt hooks that let through the actions they were written to block

Anthropic released Claude Code 2.1.294 on October 8, 2026. It fixes prompt and agent hooks phrased as instructions, which could allow exactly the actions they were meant to stop. Version 2.1.295 followed the same day with an onFailure setting that makes a crashed or timed-out command or HTTP hook block the action. Releases 2.1.292 and 2.1.293 from earlier in the week closed several permission and sandbox gaps.

Claude Code Releases (GitHub)verified