[Release] openclaw/openclaw v2026.4.25-beta.1: openclaw 2026.4.25-beta.1

OpenClaw v2026.4.25-beta.1 introduces major upgrades to text-to-speech (TTS) with multi-provider support and per-agent overrides, migrates plugin management to a cold persisted registry for faster startup, expands OpenTelemetry observability across model calls and tool execution, enhances browser automation safety with iframe-aware snapshots and CDP tuning, and hardens installation/update processes across multiple platforms. The release also adds PWA/Web Push support to the control UI and includes new provider integrations (Azure Speech, Xiaomi, Volcengine, ElevenLabs v3) alongside improved CLI tooling and diagnostics.

Key Points

• •TTS system completely redesigned with `/tts latest` read-aloud, chat-scoped auto-TTS controls (`/tts chat on|off|default`), per-agent and per-account overrides, and support for 8+ providers (Azure Speech, Xiaomi, Local CLI, Inworld, Volcengine, ElevenLabs v3)
• •Plugin registry migrated to cold persisted storage, eliminating broad manifest scans during startup and making plugin discovery, updates, and provider resolution deterministic and faster
• •OpenTelemetry coverage expanded across model calls, token usage, tool loops, harness runs, exec processes, outbound delivery, context assembly, and memory pressure with bounded low-cardinality attributes
• •Browser automation hardened with safe tab URLs, iframe-aware role snapshots, CDP readiness tuning, headless one-shot launch mode, and deeper browser doctor probes for slow hosts (e.g., Raspberry Pi)
• •Control UI and setup flows enhanced with PWA/Web Push support, Crestodian first-run repair, TUI interactive setup, context mode selection, and shorter startup greeting
• •Installation/update process hardened across Windows, macOS, Linux, Docker with bundled plugin runtime dependency management, Node service restarts, LaunchAgent token rotation, and mixed-version gateway verification
• •Plugin hooks expanded with before-agent-finalize hooks, cron jobId context, bounded native permission fingerprints, and Codex MCP relay support
• •Diagnostics infrastructure improved with signal-specific OTLP endpoint overrides, bounded telemetry exporter health diagnostics, agent harness lifecycle telemetry as spans/metrics, and W3C traceparent header propagation
• •CLI tools optimized: `openclaw plugins registry` for explicit registry inspection, `openclaw plugins list` reads cold registry by default, `openclaw browser doctor --deep` for live snapshot probing, and `openclaw browser start --headless` for one-shot launch
• •Provider discovery and capability resolution moved to cold plugin registry, eliminating runtime manifest rescans and enabling faster model listing, provider onboarding, and auth-choice setup

Workflow Diagram

Concepts

Artifacts (5)

# Per-agent TTS override
agents:
  list:
    - id: agent-1
      tts:
        provider: azure-speech
        voice: en-US-AriaNeural

# Channel and account TTS overrides
channels:
  discord:
    accounts:
      account-123:
        tts:
          provider: elevenlabs-v3
          voice: custom-voice-id

# Global TTS configuration
messages:
  tts:
    provider: default
    enabled: true

# One-shot headless browser launch
openclaw browser start --headless

# Deep browser diagnostics with live snapshot probing
openclaw browser doctor --deep

# Model listing with provider filter (optimized via cold registry)
openclaw models list --all --provider openai

# Plugin registry inspection
openclaw plugins registry

# Plugin list from cold registry
openclaw plugins list

# Plugin doctor with refresh
openclaw plugins doctor --fix --refresh

# Signal-specific OTLP endpoint overrides
diagnostics:
  otel:
    traces:
      endpoint: http://localhost:4317
    metrics:
      endpoint: http://localhost:4317
    logs:
      endpoint: http://localhost:4317

# Enable latest GenAI semantic conventions
env:
  OTEL_SEMCONV_STABILITY_OPT_IN: gen_ai_latest_experimental

# Override LLM for Discord voice channel responses
channels:
  discord:
    voice:
      model: gpt-4-turbo
      # STT and TTS remain on existing media settings

# Generic background option for image generation
openclaw infer image generate --background "blue gradient" --output-format png

# Image editing with background
openclaw infer image edit --background "white" --output-format jpeg

# OpenAI-specific alias (legacy)
openclaw infer image generate --openai-background "custom-bg"