releaseintermediate

[Release] openclaw/openclaw v2026.7.1-beta.6: openclaw 2026.7.1-beta.6

By github-actions[bot]July 13, 2026github

OpenClaw v2026.7.1-beta.6 introduces major UI/UX improvements with a redesigned Control UI featuring session-primary navigation and native macOS chat, adds support for new AI models (GPT-5.6, Claude Sonnet 5, Meta Muse Spark 1.1) and providers (Featherless, ClawRouter), implements conversational onboarding through an agent-loop setup, and enhances mobile experiences with offline caching, voice capabilities, and improved session management across iOS, Android, and Apple Watch.

Key Points

•New model support: GPT-5.6 (new default), Claude Sonnet 5, Meta Muse Spark 1.1, Mythos 5, and Featherless provider integration
•Redesigned Control UI with sessions as primary navigation, minimal searchable sidebar, compact context ring, and reasoning-effort slider
•Native macOS session browser with model/thinking pickers, slash commands, transcript export, and context usage tracking
•Conversational onboarding via Crestodian agent-loop across CLI, web, and macOS with AI-guided provider setup and masked credential prompts
•Offline mobile chat: iOS/Android pre-paint bounded session/transcript caches; Apple Watch gains full voice turns with on-device TTS fallback
•Automatic session title generation using utility-model routing with Gateway-backed groups, unread state, rename, fork, archive, and delete controls
•Telegram/Codex improvements: private `/login` pairing, `/steer` and `/tell` commands, durable turn adoption, and recovery across formatting failures
•Gateway crash-loop recovery with control-plane-safe mode, boot outcome persistence, and `EX_CONFIG` exit for fatal errors to prevent restart flapping
•ClawRouter bundled provider plugin with credential-scoped dynamic model discovery, OpenAI-compatible transports, and managed budget reporting
•Enhanced mobile features: Android agent selector in chat, iOS offline caching, Apple Watch voice dictation, Slack native thread status, and syntax-highlighted code blocks

Found this useful? Add it to a playbook for a step-by-step implementation guide.

Workflow Diagram

Start Process

Step A

Step B

Step C

Complete

Quality★★★★★

Concepts

Integrations Monitoring MCP Servers Deployment Setup & Infrastructure Releases Tool Use Updates & News

Artifacts (5)

ClawRouter Provider Configurationyamlconfig

# ClawRouter bundled provider plugin
# Credential-scoped dynamic model discovery
# OpenAI-compatible and native Anthropic/Gemini transports
# Managed budget reporting across OpenClaw usage surfaces

provider:
  type: clawrouter
  credentials:
    scope: credential-scoped
  transports:
    - openai-compatible
    - anthropic-native
    - gemini-native
  features:
    - dynamic-model-discovery
    - budget-reporting
    - usage-tracking

Crestodian Conversational Onboarding Flowyamlworkflow

# Crestodian Agent-Loop Setup Workflow
# Runs across CLI, web install, macOS app, and Gateway

steps:
  1_ai_guided_setup:
    description: "AI-guided provider setup"
    features:
      - typed-operations
      - exact-approval-binding
      - masked-credential-prompts
  
  2_model_judged_approvals:
    description: "Model-judged approvals bound to exact operations"
    features:
      - operation-validation
      - approval-binding
  
  3_deterministic_fallback:
    description: "Deterministic fallback when no model available"
    features:
      - safe-handoff
      - isolated-session-transcripts
  
  4_normal_agent_handoff:
    description: "Safe handoff to normal agent"
    features:
      - session-continuity
      - state-preservation

Telegram Codex Commandscommand

# Telegram Codex Continuity Commands

/login
  Description: Pair Codex through private login
  Usage: /login [credentials]
  Features: Private pairing, secure authentication

/steer
  Description: Steer active runs
  Usage: /steer [run_id] [direction]
  Features: Active run control, real-time steering

/tell
  Description: Send message to active run
  Usage: /tell [run_id] [message]
  Features: Message passing, run communication

Features:
  - Preserve messages when transcript acknowledgement missing
  - Adopt durable turns safely
  - Recover final sends across Telegram formatting
  - Handle flood-wait failures gracefully

Gateway Crash-Loop Recovery Configurationyamlconfig

# Gateway Crash-Loop Recovery Settings

gateway:
  recovery:
    boot_outcomes:
      persist: true
      tracking: enabled
    
    control_plane_safe_mode:
      enabled: true
      trigger: "repeated-unclean-boots"
      behavior: "safe-mode-entry"
    
    transport_activation:
      hold_until_recovery: true
    
    provider_activation:
      hold_until_recovery: true
    
    exit_codes:
      fatal_config_error: "EX_CONFIG"
      behavior: "stop-restart-flapping"
    
    systemd_integration:
      restart_policy: "on-success"
      respects_exit_codes: true
    
    launchd_integration:
      restart_policy: "on-success"
      respects_exit_codes: true
  
  startup:
    container_migrations: "before-gateway-readiness"
    legacy_state_blocking: "disabled"
    unclean_boot_recovery: "enabled"

iOS Offline Chat Cache Configurationyamlconfig

# iOS Offline Chat Cache Configuration

ios:
  offline_chat:
    caching:
      enabled: true
      scope: "per-gateway"
      bounded: true
      protected: true
    
    cache_contents:
      - recent_sessions
      - canonical_transcripts
    
    sending:
      offline_mode: "disabled"
    
    pairing_reset:
      behavior: "purge-cached-conversation-text"
    
    features:
      - pre_paint_sessions
      - pre_paint_transcripts
      - offline_availability
      - pairing_aware_cleanup