CLAUDE-CODE PUB_DATE: 2026.05.09

CLAUDE CODE 2.1.136–137 LANDS ENTERPRISE GUARDRAILS AND STABILITY FIXES (WINDOWS, MCP, OTEL)

Anthropic’s Claude Code shipped 2.1.136–137 with new enterprise controls and important reliability fixes. The [v2.1.136](https://github.com/anthropics/claude-c...

Anthropic’s Claude Code shipped 2.1.136–137 with new enterprise controls and important reliability fixes.

The v2.1.136 release adds a hard_deny switch for auto mode policies and an env var to re-enable the session quality survey for orgs piping results via OpenTelemetry. It also fixes disappearing MCP servers after /clear, stabilizes concurrent OAuth refresh across multiple servers, and adds a WSL2 clipboard fallback.

v2.1.137 quickly follows with a fix for the VS Code extension failing to activate on Windows. If you’re expanding MCP-backed workflows, the docs map is a good orientation point Claude Code docs map and this MCP-focused overview shows how to wire external systems into agent flows MCP integrations across workflows.

[ WHY_IT_MATTERS ]
01.

The Windows activation fix unblocks teams standardizing on VS Code, especially on enterprise laptops.

02.

Hard policy denies and OTel survey capture make governance and telemetry easier to enforce at scale.

[ WHAT_TO_TEST ]
  • terminal

    Upgrade to 2.1.137 and verify the VS Code extension activates on Windows and WSL2 image paste works.

  • terminal

    With multiple MCP servers using OAuth, test concurrent token refresh and confirm configs survive /clear.

[ BROWNFIELD_PERSPECTIVE ]

Legacy codebase integration strategies...

  • 01.

    Enable CLAUDE_CODE_ENABLE_FEEDBACK_SURVEY_FOR_OTEL and validate survey events flow end-to-end in your OTel pipeline.

  • 02.

    Introduce settings.autoMode.hard_deny rules for high-risk actions to codify existing security exceptions.

[ GREENFIELD_PERSPECTIVE ]

Fresh architecture paradigms...

  • 01.

    Start with auto mode hard_deny in your baseline config and design MCP integrations assuming concurrent OAuth refresh.

  • 02.

    Instrument from day one with OTel survey capture to build a feedback loop on agent quality.

Enjoying_this_story?

Get daily CLAUDE-CODE + SDLC updates.

  • Practical tactics you can ship tomorrow
  • Tooling, workflows, and architecture notes
  • One short email each weekday

FREE_FOREVER. TERMINATE_ANYTIME. View an example issue.

GET_DAILY_EMAIL
AI + SDLC // 5 MIN DAILY