Claude Code 2.1.162 tightens permissions, resolves startup hangs

// 58d agoPRODUCT UPDATE

Claude Code 2.1.162 tightens permissions, resolves startup hangs

Anthropic has released Claude Code version 2.1.162, introducing stricter WebFetch permission overrides and resolving a silent startup hang when config directories are unwritable. The update also removes several CLI environment variables to clean up the configuration surface.

// ANALYSIS

Tightening execution boundaries and improving resilience in constrained environments are essential steps for maturing terminal-based agentic tools.

* WebFetch overrides ensure that custom user rules always take precedence over default preapproved domains, preventing potential data exfiltration.

* Reverting to in-memory configuration solves permissions-related environment issues, preventing silent startup failures in CI pipelines or containerized environments.

* The removal of numerous environment variables suggests code refactoring and simplification of configuration options as the CLI approaches stability.

// TAGS

claude-codeanthropicagentdevtoolcli

DISCOVERED

58d ago

2026-06-03

PUBLISHED

58d ago

2026-06-03

RELEVANCE

8/ 10

AUTHOR

ClaudeCodeLog

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

MODEL48m ago

DeepSeek-V4-Flash-High excels at low-cost frontend coding

AI researcher Elvis Saravia (@omarsar0) highlighted the impressive front-end development capabilities of DeepSeek-V4-Flash-High during recent testing. He noted that the model's output quality was high enough to prompt a double-check of which model was actively being used, praising its performance-to-price ratio.

TUTORIAL1h ago

DAIR.AI offers harness engineering, evals training

DAIR.AI emphasizes harness engineering and model evaluations as essential skills for building production-grade AI applications. The platform is releasing educational resources and courses focused on evaluation harnesses and systematic testing.

TUTORIAL1h ago

Dual Blackwell GPUs run 167 GB DeepSeek-V4 FP8

A developer shared a deployment recipe for running the official FP8 version of DeepSeek-V4-Flash-0731 alongside DSpark speculative decoding on a dual NVIDIA RTX PRO 6000 Blackwell (SM120) GPU rig. Requiring approximately 167 GB of VRAM, the model fits cleanly across the system's combined 192 GB VRAM capacity (2× 96 GB) without offloading or truncation.