// event · cli

Claude Code v2.1.283: prompt-id gateway hints, model allow/deny lists, prompt audit

cli

vendor: Anthropicpublished: 17d agoconfidence: 0.85id: 9ec9268f-d080-476b-8d65-819933b76602

Claude Code CLI 2.1.283 ships gateway hint headers with an `x-claude-code-prompt-id` for grouping requests per user prompt, new `availableModelsMatch=exact` and `deniedModels` managed settings to lock down model access, a `/doctor prompt-audit` command that scans CLAUDE.md/skills/agents/commands for outdated prompting patterns, MCP tool/WebFetch/WebSearch output capture in OpenTelemetry spans, and many MCP, plugin, sandbox, and vim-mode fixes.

capability delta

An agent operator can now pin exact allowed model versions, block specific models, audit prompt files for outdated patterns, and correlate LLM gateway requests to a single user prompt via a new header.

curator note

Point release but ships several agent-relevant capabilities: model allow/deny governance, prompt audit tooling, and OTel tool-output telemetry.

// install

npm i -g @anthropic-ai/claude-code@2.1.283

npm · @anthropic-ai/claude-code @ 2.1.283

// example

bash

npm i -g @anthropic-ai/claude-code@2.1.283

# Enable gateway hint headers so an LLM gateway can group requests per prompt
export CLAUDE_CODE_GATEWAY_HINT_HEADERS=1

# Audit CLAUDE.md, skills, agents, and commands for outdated prompt patterns
claude /doctor prompt-audit

# Capture MCP tool / WebFetch / WebSearch outputs in OTel spans
export OTEL_LOG_TOOL_CONTENT=1
claude

// links

claude-codeclimcpopentelemetrypluginsgovernance

source: github_release · https://github.com/anthropics/claude-code/releases/tag/v2.1.283