Get started

Maturity taxonomy

Maturity taxonomy

This page explains the product areas and capability groups behind the maturity scorecard.

Maturity levels

Level	Label	Meaning	Promotion bar
`M0`	Planned	Direction is known, but no supported user path exists.	Design issue, owner, and target surface exist.
`M1`	Experimental	Implemented behind caveats, flags, source builds, or maintainer-only flows.	Maintainer can run the scenario from current main.
`M2`	Alpha	Real users can try it, but breaking changes and incomplete UX are expected.	Documented setup, basic tests, known caveats, and at least one real-environment proof.
`M3`	Beta	Public path exists and the main workflow is usable with bounded caveats.	Install/update docs, regression tests, support runbook, and successful scenario proof across the expected environment.
`M4`	Stable	Recommended path for normal users. Failures are treated as regressions.	Release gate, doctor/troubleshooting path, broad docs, and repeated real-world proof.
`M5`	Lovable	Polished, delightful, well-instrumented, and competitive with the best comparable workflow.	Stable plus user scorecard pass across representative users.

Product areas

Core surfaces

Gateway runtime
CLI
Plugins
Agent Runtime
Session, memory, and context engine
Channel framework
Security, auth, pairing, and secrets
Observability
Automation: cron, hooks, tasks, polling
Media understanding and media generation
Voice and realtime talk
Gateway Web App
TUI
ClawHub
OpenClaw App SDK

Platform surfaces

macOS Gateway host
macOS companion app
Linux Gateway host
Linux companion app
Windows via WSL2
Native Windows
Native Windows companion app
Android app
iOS app
watchOS companion surfaces
Raspberry Pi and small Linux devices
Docker and Podman hosting
Kubernetes hosting
Nix install path

Channel surfaces

Discord
Telegram
WhatsApp
Slack
iMessage and BlueBubbles
Signal
Google Chat
Matrix
Microsoft Teams
Mattermost, LINE, IRC, Nextcloud Talk, Nostr, Twitch, Tlon, Synology Chat
Feishu, QQ Bot, WeChat, Yuanbao, Zalo, Zalo Personal, regional channels
Voice Call channel

Provider and tool surfaces

OpenAI and Codex provider path
Anthropic provider path
Google provider path
OpenRouter provider path
Local model providers: Ollama, vLLM, SGLang, LM Studio
Long-tail hosted providers
Web search tools
Browser automation, exec, and sandbox tools
Image, video, and music generation tools

Details

Core

Gateway runtime

Level: M4 Stable
Rationale: Core architecture, auth, pairing, protocol docs, daemon docs, and CLI runbooks are broad and current.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Approvals and Remote Execution	6	Protocol, Index	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
HTTP APIs	4	Index, Openai Http Api, Openresponses Http Api, Tools Invoke Http Api, Hooks, Index	`Experimental (25%)`	`Stable (90%)`	`Stable (90%)`	Yes
Hosted Web Surface	4	Index, Architecture, Control Ui, Webchat, Canvas	`Experimental (0%)`	`Stable (89%)`	`Stable (90%)`	Yes
Gateway RPC APIs and Events	20	Protocol, Index, Architecture	`Experimental (0%)`	`Stable (90%)`	`Stable (90%)`	Yes
Device Auth and Pairing	10	Protocol, Pairing, Index	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Network Access and Discovery	6	Index, Discovery, Protocol	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Nodes and Remote Capabilities	8	Protocol, Architecture, Index	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	No
Health, Diagnostics, and Repair	7	Index, Diagnostics, Doctor	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Protocol Compatibility	7	Protocol, Architecture, Typebox, Bridge Protocol	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Roles and Permissions	5	Protocol, Index	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Gateway Lifecycle	7	Index, Architecture	`Experimental (0%)`	`Stable (90%)`	`Stable (90%)`	Yes
Security Controls	6	Index, Protocol, Discovery	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
WebSocket Connection	8	Protocol, Architecture	`Experimental (13%)`	`Stable (90%)`	`Stable (90%)`	Yes

CLI

Level: M4 Stable
Rationale: Normal setup and repair paths are documented across install, CLI, and gateway docs. Platform-specific Windows paths are tracked in the Windows via WSL2 and Native Windows rows.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
CLI Setup	6	Index, Installer, Node, Updating	`Experimental (17%)`	`Stable (89%)`	`Stable (90%)`	Yes
Onboarding and Auth Setup	5	Onboard, Configure, Onboarding Overview	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Plugin and Channel Setup	5	Onboard, Plugins, Channels	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	No
Gateway Service Management	5	Gateway, Updating, Troubleshooting	`Experimental (0%)`	`Stable (87%)`	`Stable (90%)`	Yes
CLI Observability	5	Status, Health, Logs, Diagnostics	`Experimental (0%)`	`Stable (89%)`	`Stable (90%)`	Yes
Doctor	10	Doctor, Doctor, Secrets, Troubleshooting	`Experimental (0%)`	`Stable (89%)`	`Stable (90%)`	Yes
Updates and Upgrades	5	Updating, Update, Troubleshooting	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes

Plugins

Level: M3 Beta
Rationale: Broad docs and strong internal runtime evidence exist across manifests, discovery, loading, provider/tool architecture, and approval boundaries. Keep the row at beta until public SDK API/subpaths and external distribution proof are stronger.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Authoring and Packaging plugins	8	Building Plugins, Sdk Overview, Sdk Entrypoints, Sdk Subpaths, Manifest, Reference	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Bundled plugins	5	Plugin Inventory, Plugins, Architecture Internals	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Canvas plugin	6	Canvas, Canvas, Configuration Reference	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Installing and running plugins	6	Architecture, Architecture Internals, Plugins	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Channel plugins	5	Sdk Channel Plugins, Sdk Channel Inbound, Sdk Channel Outbound	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Provider and tool plugins	6	Sdk Provider Plugins, Tool Plugins, Adding Capabilities	`Experimental (17%)`	`Beta (79%)`	`Beta (79%)`	Yes
Plugin approvals	6	Plugin Permission Requests, Exec Approvals, Sdk Channel Plugins	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Publishing plugins	6	Plugins, Compatibility, Publishing	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Testing plugins	6	Sdk Testing, Sdk Setup, Codex Harness	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No

Agent Runtime

Level: M3 Beta
Rationale: Main loop, models, provider routing, and tool streaming are first-class, but provider behavior shifts weekly and needs scenario proof per release.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Agent Turn Execution	3	Agent Loop, Agent, Agent Runtimes	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
External Runtimes and Subagents	4	Agent Runtimes, Anthropic, Google, Subagents	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Hosted Provider Execution	5	Openai, Anthropic, Google, Models	`Experimental (20%)`	`Beta (79%)`	`Beta (79%)`	Yes
Local and Self-hosted Providers	5	Ollama, Models, Agent	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Model and Runtime Selection	4	Models, Models, Openai, Agent Runtimes	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Provider Auth	10	Models, Agent, Models, Openai, Anthropic, Google, Subagents	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Streaming and Progress	2	Streaming, Agent Loop	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Tool Calls and Response Handling	3	Agent Loop, Ollama	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Tool Execution Controls	6	Sandbox Vs Tool Policy Vs Elevated, Agent Loop, Subagents	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes

Session, memory, and context engine

Level: M3 Beta
Rationale: Strong docs and active implementation. Maturity depends on transcript durability, compaction quality, and cross-client parity.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
CLI Session and Transcript Management	2	Session, Session Management Compaction, Sessions	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Token Management	3	Compaction, Context, Session Management Compaction	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Context Engine	2	Context, Context Engine, Codex Context Engine Harness	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Cross-client History and Session Parity	2	Webchat, Android, Channel Routing	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Diagnostics, Maintenance, and Recovery	3	Diagnostics, Session Management Compaction, Flags	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Core Prompts and Context	2	Context, Transcript Hygiene, Discord	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Memory	5	Memory Config, Memory Qmd, Memory, Discord	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Session Routing	2	Session, Channel Routing, Discord	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Transcript Persistence	2	Session Management Compaction, Transcript Hygiene	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes

Channel framework

Level: M3 Beta
Rationale: Many channels share Gateway delivery and routing contracts, but channel behavior varies by upstream API and account-policy constraints.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Actions Commands and Approvals	5	Groups, Discord, Googlechat, Signal, Matrix	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Channel Setup	5	Index, Pairing, Troubleshooting, Sdk Channel Plugins	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Group Thread and Ambient Room Behavior	5	Groups, Group Messages, Ambient Room Events, Broadcast Groups, Discord	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Inbound Access and Identity Gates	5	Access Groups, Groups, Discord, Line	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Media Attachments and Rich Channel Data	4	Line, Signal, Googlechat, Matrix, Discord	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Outbound Delivery and Reply Pipeline	4	Groups, Ambient Room Events, Discord, Matrix, Config Channels	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Conversation Routing and Delivery	10	Channel Routing, Groups, Discord, Matrix, Troubleshooting, Configuration Reference	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Status Health and Operator Controls	4	Health, Configuration Reference, Troubleshooting, Discord	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes

Security, auth, pairing, and secrets

Level: M3 Beta
Rationale: Good docs and hardening surfaces exist. Promote after regular upgrade/security scenario runs prove no setup regressions.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Approval Policy and Tool Safeguards	2	Exec Approvals, Approvals, Plugin Permission Requests, Audit Checks	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Gateway Auth and Remote Access	9	Index, Exposure Runbook, Trusted Proxy Auth, Tailscale, Remote, Configuration Reference, Gateway, Doctor, Control Ui, Browser Control, Audit Checks	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Channel Access Control	3	Pairing, Telegram, Access Groups, Audit Checks	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Device and Node Pairing	11	Protocol, Devices, Pairing, Pairing, Operator Scopes, Control Ui, Webchat, Approvals	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Plugin Trust	2	Manifest, Plugin Permission Requests, Manage Plugins, Audit Checks	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Credential and Secret Hygiene	5	Authentication, Models, Openai, Oauth, Secrets, Secrets, Secretref Credential Surface, Audit Checks	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes

Observability

Level: M3 Beta
Rationale: OTel, Prometheus, logging, and diagnostics docs exist. Needs a public "what operators should look at first" maturity pass.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Health and Repair	12	Health, Telegram, Doctor, Doctor, Sdk Subpaths, Health, Protocol	`Experimental (8%)`	`Beta (79%)`	`Beta (79%)`	Yes
Logging	5	Logging, Logging, Logs	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes
Diagnostic Collection	8	Diagnostics, Health, Codex Harness, Protocol	`Experimental (13%)`	`Beta (79%)`	`Beta (79%)`	No
Telemetry Export	13	Hooks, Opentelemetry, Logging, Sdk Subpaths, Diagnostics Otel, Prometheus, Diagnostics Prometheus	`Experimental (8%)`	`Beta (79%)`	`Beta (79%)`	No
Session Diagnostics	4	Opentelemetry, Prometheus, Diagnostics, Protocol	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes

Automation: cron, hooks, tasks, polling

Level: M3 Beta
Rationale: Documented and usable, but scenario proof should cover unattended delivery, retries, and failure visibility.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Cron Jobs	15	Cron Jobs, Cron, Protocol, Tasks, Discord	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Event Ingress	15	Telegram, Zalo, Troubleshooting, Imessage From Bluebubbles, Gmail Pubsub Integration, Gmail Pubsub, Webhooks, Webhooks, Webhook	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Automation Hooks	11	Hooks, Hooks, Hooks, Plugin Permission Requests, Sdk Subpaths	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Background Tasks and Flows	10	Tasks, Index, Tasks, Taskflow, Sdk Runtime	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Heartbeat	5	Index, Heartbeat, Commitments	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Polling Controls	10	Poll, Message, Telegram, Msteams, Background Process	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No

Media understanding and media generation

Level: M2 Alpha
Rationale: Broad capability surface exists, but provider variance, file limits, and node/app parity make this not stable yet.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Media Intake and Access	8	Media Overview, Media Understanding, Secure File Operations, Pdf, Image Generation, Qr, Line, Whatsapp	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Channel Media Handling	5	Images, Media Overview, Discord	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Media Configuration	1	Media Overview, Image Generation, Manifest, Codex Harness	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Text-to-Speech Delivery	2	Tts, Media Overview, Discord	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Media Understanding	12	Audio, Media Understanding, Media Overview, Whatsapp, Images, Infer, Pdf	`Experimental (0%)`	`Alpha (69%)`	`Alpha (69%)`	No
Media Generation	17	Image Generation, Media Overview, Skills, Music Generation, Video Generation	`Experimental (6%)`	`Alpha (69%)`	`Alpha (69%)`	No

Voice and realtime talk

Level: M2 Alpha
Rationale: Multiple implementations exist across Control UI, apps, and providers. Needs latency, failure-mode, and setup scorecards before beta.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Talk Providers	7	Openai, Google, Sdk Provider Plugins, Talk, Control Ui	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Realtime Talk Sessions	11	Talk, Control Ui	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Speech and Transcription	5	Talk, Openai, Google	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Native App Talk	4	Talk, Voicewake	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Voice Wake and Routing	4	Voicewake, Voicewake, Voice Overlay	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Talk Observability	5	Control Ui, Voice Overlay, Talk	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No

Gateway Web App

Level: M3 Beta
Rationale: Web UI is documented with pairing, chat, PWA, Talk, push, and remote Gateway flows. Promote after cross-browser and mobile-PWA scorecards.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Browser Realtime Talk	5	Control Ui, Protocol, Talk	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Browser Access and Trust	5	Control Ui, Dashboard, Tailscale, Remote	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Configuration	5	Control Ui, Configuration	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Browser UI	10	Control Ui, Index, Dashboard, Protocol	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
WebChat Conversations	15	Control Ui, Webchat, Getting Started, Channel Routing, Secure File Operations	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No
Operator Console	10	Control Ui, Health, Protocol, Dashboard	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	No

TUI

Level: M2 Alpha
Rationale: Present in docs and source, but less visible as a primary user workflow. Needs explicit scenario definition.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Runtime Modes	14	Tui, Tui, Index	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Input and Commands	8	Tui	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Session Management	3	Tui, Sessions	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Local Shell Execution	4	Tui, Tui	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Rendering and Output Safety	4	Tui, Qr, Logs, Completion	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No

ClawHub

Level: M2 Alpha
Rationale: Public docs and ecosystem concept exist. Needs install, trust, update, rollback, and compatibility scorecards.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Publishing	7	Publishing, Creating Skills, Community	`Experimental (0%)`	`Alpha (54%)`	`Alpha (55%)`	No
Catalog Discovery	5	Plugin, Plugins, Skills, Skills, Community	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Compatibility and Trust	12	Plugin, Plugins, Compatibility, Plugin Inventory, Publishing, Skills, Skills Config	`Experimental (0%)`	`Alpha (55%)`	`Alpha (56%)`	No
Plugin Lifecycle and Health	26	Plugin, Plugins, Skills, Skills, Protocol, Bundles, Dependency Resolution	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No

OpenClaw App SDK

Level: M2 Alpha
Rationale: OpenClaw App SDK is a distinct external app contract separate from Gateway runtime and Plugin SDK. Current scoring shows a real @openclaw/sdk path with gaps around public packaging, auto-discovery, approvals, helpers, and compatibility.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Client API	4	Openclaw Sdk, Openclaw Sdk Api Design	`Experimental (0%)`	`Alpha (51%)`	`Alpha (50%)`	No
Gateway Access	5	Openclaw Sdk, Openclaw Sdk Api Design, Protocol, Index	`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No
Agent Conversations	6	Openclaw Sdk, Openclaw Sdk Api Design, Protocol	`Experimental (0%)`	`Alpha (52%)`	`Alpha (52%)`	No
Events and Approvals	5	Openclaw Sdk, Openclaw Sdk Api Design, Protocol	`Experimental (0%)`	`Alpha (52%)`	`Alpha (52%)`	No
Resource Helpers	5	Openclaw Sdk, Openclaw Sdk Api Design	`Experimental (0%)`	`Alpha (62%)`	`Alpha (53%)`	No
Compatibility	5	Openclaw Sdk Api Design, Typebox, Protocol	`Experimental (0%)`	`Alpha (54%)`	`Alpha (55%)`	No

Platform

macOS Gateway host

Level: M4 Stable
Rationale: LaunchAgent service path, local/remote Gateway modes, CLI install, and app integration are documented.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
CLI Setup	4	Macos, Bundled Gateway, Installer, Node	`Experimental (0%)`	`Beta (74%)`	`Stable (88%)`	No
Local Gateway Integration	9	Macos, Bundled Gateway, Remote, Index, Gateway, Bonjour	`Experimental (0%)`	`Beta (74%)`	`Stable (88%)`	No
Remote Gateway Mode	5	Remote, Remote, Tailscale	`Experimental (0%)`	`Beta (74%)`	`Stable (88%)`	No
Gateway Service Lifecycle	10	Macos, Bundled Gateway, Gateway, Index, Update, Updating, Uninstall, Troubleshooting	`Experimental (0%)`	`Beta (74%)`	`Stable (88%)`	No
Diagnostics and Observability	4	Bundled Gateway, Macos, Gateway, Doctor, Troubleshooting	`Experimental (0%)`	`Beta (74%)`	`Stable (88%)`	No
Permissions and Native Capabilities	4	Macos, Remote	`Experimental (0%)`	`Beta (74%)`	`Stable (88%)`	No
Profiles and Isolation	5	Multiple Gateways, Index, Gateway	`Experimental (0%)`	`Beta (74%)`	`Stable (88%)`	No

macOS companion app

Level: M3 Beta
Rationale: Rich menu bar app, permissions, node mode, Canvas, voice wake, WebChat, and remote mode exist. Still fast-moving enough to avoid Stable.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Canvas	4	Canvas, Macos, Webchat	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Local Setup	7	Bundled Gateway, Macos, Child Process, Dev Setup	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Status and Settings	5	Menu Bar, Icon, Macos, Health, Logging, Remote	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Native Capabilities	5	Macos, Xpc, Permissions, Signing, Peekaboo	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Remote Connections	3	Remote, Macos, Remote	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Voice and Talk	3	Voicewake, Voice Overlay, Talk, Macos	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
WebChat	3	Webchat, Macos, Webchat	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Remote WebChat	5	Webchat, Remote, Remote	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No

Linux Gateway host

Level: M4 Stable
Rationale: Node runtime is recommended, systemd user service is documented, and VPS/container guidance is broad.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Host Setup and Updates	4	Index, Updating, Linux, Index	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Gateway Runtime and Service Control	6	Index, Gateway, Linux, Vps	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Remote Access and Security	6	Remote, Tailscale, Exposure Runbook, Authentication, Secrets	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Diagnostics and Repair	4	Status, Logs, Doctor, Diagnostics, Index	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	Yes
Deployment Targets	3	Vps, Docker, Hetzner, Digitalocean, Kubernetes, Podman	`Experimental (0%)`	`Beta (75%)`	`Stable (89%)`	No

Linux companion app

Level: M0 Planned
Rationale: Docs say native Linux companion apps are planned; Gateway is the supported Linux path today.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
App Distribution	3	Linux, Index, Index	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Gateway Connectivity	4	Linux, Index, Pairing, Remote	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Chat and Sessions	3	Linux, Protocol, Webchat	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Desktop Capabilities	9	Linux, Exec Approvals, Secrets, Index, Exec, Talk, Camera	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Status and Diagnostics	7	Linux, Openclaw, Doctor	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No

Windows via WSL2

Level: M3 Beta
Rationale: Recommended Windows path with systemd/user-service guidance and boot-chain docs. Promote after repeated install/update scorecards.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
WSL Setup	6	Windows, Getting Started	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	Yes
CLI	8	Windows, Getting Started, Updating, Onboard, Doctor, Status, Logs	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	Yes
Gateway Service Lifecycle	10	Windows, Index, Doctor	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	Yes
Gateway Access and Exposure	11	Authentication, Secrets, Remote, Exposure Runbook, Windows	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	Yes
Diagnostics and Repair	6	Windows, Status, Logs, Doctor, Doctor	`Experimental (17%)`	`Beta (79%)`	`Beta (79%)`	Yes
Browser and Control UI	6	Browser Wsl2 Windows Remote Cdp Troubleshooting, Browser, Control Ui	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	No

Native Windows

Level: M2 Alpha
Rationale: Core CLI/Gateway flows work, but docs still recommend WSL2 for the full experience and list native caveats.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
CLI	9	Index, Installer, Windows, Getting Started, Onboard	`Experimental (0%)`	`Alpha (54%)`	`Alpha (64%)`	Yes
Gateway Management	11	Windows, Index, Gateway, Doctor	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Networking	4	Windows, Index, Gateway	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Updates	4	Updating, Ci	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No

Native Windows companion app

Level: M0 Planned
Rationale: Planned only.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Installation and Updates	4	Windows, Index	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Gateway Connection	3	Windows, Index, Pairing, Remote	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Chat Sessions	2	Windows, Protocol	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Status and Repair	5	Windows, Doctor, Index	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No
Desktop Tools and Permissions	10	Windows, Index, Exec, Exec Approvals, Index	`Experimental (0%)`	`Experimental (19%)`	`Experimental (21%)`	No

Android app

Level: M2 Alpha
Rationale: Public Google Play path exists, but app docs still describe the rebuild as extremely alpha and call out release hardening work.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Media Capture	1	Android, Camera	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Mobile Chat	1	Android	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Connection Setup	1	Android, Bonjour, Pairing	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Distribution	3	Android	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Settings	1	Android	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Voice	1	Android, Talk	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Device Runtime	2	Android, Troubleshooting, Protocol	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No

iOS app

Level: M1 Experimental
Rationale: Internal preview / super-alpha. TestFlight and relay-backed push flows exist, but no public distribution yet.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Media and Sharing	1	Ios, Camera	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Canvas and Screen	1	Ios, Canvas	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Chat and Sessions	1	Ios, Webchat, Protocol	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Gateway Setup and Diagnostics	7	Ios, Pairing	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Distribution	1	Ios	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Device Commands	2	Ios, Protocol	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Notifications and Background	1	Ios, Configuration	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Voice	1	Ios, Talk	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No

watchOS companion surfaces

Level: M1 Experimental
Rationale: Source has Watch app/extension surfaces; public docs do not yet present this as a user feature.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Delivery and Recovery	7	Ios	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Exec Approvals	3	Exec Approvals, Ios	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Distribution and Support	6	Ios	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Notifications and Replies	7	Ios	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Watch App UI	3	Ios	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No

Raspberry Pi and small Linux devices

Level: M3 Beta
Rationale: Platform docs exist and Gateway path is Linux-based. Needs hardware-specific release smoke proof to move higher.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Setup and Compatibility	12	Raspberry Pi, Index, Faq First Run, Faq, Linux, Installer	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	No
Remote Access and Auth	9	Raspberry Pi, Authentication, Secrets, Pairing, Devices, Remote, Tailscale	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	No
Gateway Runtime	10	Index, Gateway, Raspberry Pi, Linux, Vps	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	No
Performance and Diagnostics	5	Raspberry Pi, Linux, Health, Diagnostics	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	No

Docker and Podman hosting

Level: M3 Beta
Rationale: Install docs exist and are common deployment paths. Promote after recurring release smoke captures upgrade and volume behavior.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Container Setup	6	Docker, Podman	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Container Operations	11	Podman, Docker Vm Runtime, Docker, Hetzner, Hostinger	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Image Release and Validation	5	Docker, Docker Vm Runtime, Full Release Validation	`Experimental (20%)`	`Beta (79%)`	`Beta (79%)`	No
Agent Sandbox and Tooling	3	Docker, Docker Vm Runtime	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No

Kubernetes hosting

Level: M2 Alpha
Rationale: Kubernetes hosting is a distinct Kustomize-based cluster deployment path. Current scoring shows a real minimal deployment path with gaps around Kubernetes-specific CI, ingress/TLS/NetworkPolicy packaging, backup/restore, and production exposure hardening.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Deployment Setup	5	Kubernetes, Index	`Experimental (0%)`	`Alpha (55%)`	`Alpha (61%)`	No
Configuration and Secrets	5	Kubernetes, Secrets, Environment	`Experimental (0%)`	`Alpha (55%)`	`Alpha (61%)`	No
Access and Exposure	5	Kubernetes, Authentication, Remote, Exposure Runbook	`Experimental (0%)`	`Alpha (55%)`	`Alpha (61%)`	No
Cluster Lifecycle	5	Kubernetes, Index	`Experimental (0%)`	`Alpha (55%)`	`Alpha (61%)`	No

Nix install path

Level: M1 Experimental
Rationale: Optional install flow. Needs clearer support promise before alpha/beta promotion.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Install Handoff	4	Nix, Index, Docs Directory	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Plugin Lifecycle	4	Manage Plugins, Plugin, Nix	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Activation and App UX	7	Nix	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Config and State	7	Nix, Setup, Environment	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Service Runtime and Guards	8	Nix, Setup, Doctor, Update	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No

Channel

Discord

Level: M4 Stable
Rationale: Deep docs and broad feature coverage. Voice/delegation paths should stay separately scored as beta/alpha.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	10	Discord, Discord, Fly, Slash Commands, Health, Channels, Config Channels	`Experimental (0%)`	`Beta (73%)`	`Stable (87%)`	Yes
Access and Identity	6	Discord, Pairing, Access Groups, Groups	`Experimental (0%)`	`Beta (73%)`	`Stable (87%)`	Yes
Conversation Routing and Delivery	12	Discord, Channel Routing, Groups, Access Groups, Acp Agents, Subagents	`Experimental (0%)`	`Beta (73%)`	`Stable (87%)`	Yes
Media and Rich Content	1	Discord	`Experimental (0%)`	`Beta (73%)`	`Stable (87%)`	Yes
Native Controls and Approvals	5	Discord, Slash Commands	`Experimental (0%)`	`Beta (73%)`	`Stable (87%)`	No
Realtime Voice and Calls	5	Discord, Openai, Elevenlabs, Qa E2e Automation, Config Channels	`Experimental (0%)`	`Beta (73%)`	`Stable (87%)`	No

Telegram

Level: M3 Beta
Rationale: Core channel is mature enough for regular use, but high-variance UX and media edge cases need recurring scenario proof.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	10	Telegram, Config Channels, Channels	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Access and Identity	10	Telegram, Pairing, Access Groups, Groups, Multi Agent	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Conversation Routing and Delivery	1	Telegram, Groups, Multi Agent	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Media and Rich Content	1	Telegram, Location	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Native Controls and Approvals	9	Telegram, Exec Approvals, Reactions	`Experimental (0%)`	`Beta (77%)`	`Beta (79%)`	Yes

WhatsApp

Level: M3 Beta
Rationale: Core path is important and documented; upstream Baileys/session volatility keeps it below Stable.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	5	Whatsapp, Config Channels, Whatsapp, Qa E2e Automation, Doctor	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Access and Identity	7	Whatsapp, Config Channels, Qa E2e Automation, Pairing	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Conversation Routing and Delivery	4	Whatsapp, Group Messages	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Media and Rich Content	2	Whatsapp	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Native Controls and Approvals	2	Whatsapp	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No

Slack

Level: M3 Beta
Rationale: First-class channel docs and routing surface. Needs workspace install/admin scenario scorecards.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	10	Slack, Slack, Secrets, Qa E2e Automation, Troubleshooting	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Access and Identity	1	Slack, Pairing	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Conversation Routing and Delivery	5	Slack, Bot Loop Protection, Pairing	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Media and Rich Content	1	Slack, Qa E2e Automation	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes
Native Controls and Approvals	8	Slack, Slash Commands, Exec Approvals	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	Yes

iMessage and BlueBubbles

Level: M3 Beta
Rationale: Supported iMessage runs through imsg on a signed-in macOS Messages host; legacy BlueBubbles configs require migration. Keep macOS permissions, SSH wrapper, SIP/private API, and migration caveats visible.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	11	Bluebubbles Imessage, Imessage From Bluebubbles, Config Channels, Imessage	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Access and Identity	6	Imessage, Imessage From Bluebubbles, Config Channels	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Conversation Routing and Delivery	4	Imessage	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Media and Rich Content	7	Imessage, Imessage From Bluebubbles, Config Channels	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Native Controls and Approvals	3	Imessage	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No

Signal

Level: M2 Alpha
Rationale: Supported channel docs exist; needs stronger install and reconnect proof.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	7	Signal, Signal	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Access and Identity	6	Signal	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Conversation Routing and Delivery	1	Signal	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Media and Rich Content	7	Signal	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Native Controls and Approvals	3	Signal	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No

Google Chat

Level: M2 Alpha
Rationale: Documented channel, but enterprise/admin setup raises maturity risk.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	16	Googlechat, Googlechat, Config Channels, Wizard Cli Reference, Secrets, Secretref Credential Surface, Health, Plugin Inventory, Index	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Access and Identity	11	Googlechat, Pairing, Access Groups, Config Channels, Bot Loop Protection, Channel Routing	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Conversation Routing and Delivery	1	Googlechat, Bot Loop Protection, Access Groups, Channel Routing	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Media and Rich Content	1	Googlechat, Message, Media Understanding, Secretref Credential Surface	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Native Controls and Approvals	16	Googlechat, Message, Media Understanding, Secretref Credential Surface, Reactions, Slash Commands, Config Agents, Message Lifecycle Refactor	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No

Matrix

Level: M2 Alpha
Rationale: Supported via bundled plugin. Needs bridge, auth, and room lifecycle scorecards.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	5	Matrix, Matrix Migration	`Experimental (0%)`	`Alpha (60%)`	`Alpha (67%)`	No
Access and Identity	7	Matrix, Groups, Bot Loop Protection	`Experimental (0%)`	`Alpha (60%)`	`Alpha (67%)`	No
Conversation Routing and Delivery	1	Matrix	`Experimental (0%)`	`Alpha (60%)`	`Alpha (67%)`	No
Media and Rich Content	1	Matrix	`Experimental (0%)`	`Alpha (60%)`	`Alpha (67%)`	No
Native Controls and Approvals	6	Matrix	`Experimental (0%)`	`Alpha (60%)`	`Alpha (67%)`	No
Encryption and Verification	3	Matrix, Matrix Migration	`Experimental (0%)`	`Alpha (60%)`	`Alpha (67%)`	No

Microsoft Teams

Level: M2 Alpha
Rationale: Enterprise auth/admin flows need explicit scenario proof.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	9	Msteams, Msteams, Config Channels, Health	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Access and Identity	9	Msteams, Pairing, Access Groups	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Conversation Routing and Delivery	5	Msteams, Groups, Channel Routing	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Media and Rich Content	5	Msteams	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No
Native Controls and Approvals	5	Msteams, Exec Approvals Advanced	`Experimental (0%)`	`Alpha (59%)`	`Alpha (66%)`	No

Mattermost, LINE, IRC, Nextcloud Talk, Nostr, Twitch, Tlon, Synology Chat

Level: M2 Alpha
Rationale: Supported surfaces exist, but maturity likely varies by upstream and maintainer coverage. Score individually later.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	1		`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No
Access and Identity	1		`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No
Conversation Routing and Delivery	1		`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No
Media and Rich Content	1		`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No

Feishu, QQ Bot, WeChat, Yuanbao, Zalo, Zalo Personal, regional channels

Level: M2 Alpha
Rationale: Important regional coverage, but public support level should be calibrated per account type, upstream approval, and maintainer proof.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	6	Index, Pairing, Feishu, Architecture Internals	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Access and Identity	1		`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No
Conversation Routing and Delivery	1		`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No
Media and Rich Content	1		`Experimental (0%)`	`Alpha (53%)`	`Alpha (54%)`	No

Voice Call channel

Level: M1 Experimental
Rationale: Optional/plugin path with complex realtime behavior. Needs scenario scorecard before public beta.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Channel Setup and Operations	2	Voicecall, Voice Call, Protocol	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Access and Identity	1	Voice Call, Voicecall	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Conversation Routing and Delivery	1	Voice Call	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Media and Rich Content	2	Voice Call, Plugin Inventory	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No
Realtime Voice and Calls	2	Voice Call	`Experimental (0%)`	`Experimental (41%)`	`Experimental (44%)`	No

Provider and tool

OpenAI and Codex provider path

Level: M3 Beta
Rationale: Deep docs, OAuth/subscription path, realtime voice, image, and compatibility behavior. Provider churn keeps this from Stable without release-scorecard proof.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Model and Auth	6	Openai, Codex Harness, Models, Oauth, Codex Harness Reference, Auth Monitoring	`Experimental (17%)`	`Beta (79%)`	`Beta (79%)`	Yes
Responses and Tool Compatibility	4	Openai, Openresponses Http Api, Openai Http Api, Codex Native Plugins	`Experimental (25%)`	`Beta (79%)`	`Beta (79%)`	Yes
Native Codex Harness	2	Codex Harness, Codex Harness Runtime, Codex Harness Reference, Codex Native Plugins	`Experimental (0%)`	`Beta (79%)`	`Beta (79%)`	Yes
Image and Multimodal Input	2	Openai, Image Generation, Images	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	No
Voice and Realtime Audio	2	Openai, Discord, Voice Call	`Experimental (0%)`	`Alpha (67%)`	`Beta (79%)`	No

Anthropic provider path

Level: M3 Beta
Rationale: First-class model provider. Needs recurring auth/catalog/tool-call scenario proof.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Provider Auth and Recovery	9	Anthropic, Doctor, Configuration Examples, Troubleshooting, Prompt Caching	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Model and Runtime Selection	10	Anthropic, Config Agents, Models, Cli Backends	`Experimental (0%)`	`Beta (78%)`	`Beta (79%)`	No
Request Transport and Turn Semantics	10	Anthropic, Prompt Caching, Troubleshooting, Cli Backends, Model Providers	`Experimental (0%)`	`Beta (77%)`	`Beta (79%)`	No
Prompt Cache and Context	5	Anthropic, Prompt Caching, Troubleshooting, Heartbeat	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Media Inputs	4	Anthropic, Config Agents	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No

Google provider path

Level: M3 Beta
Rationale: First-class provider with model and realtime surfaces. Needs separate Live/Talk scoring.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Provider Setup and Credentials	10	Google, Model Providers	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Model Routing and Endpoints	10	Google, Model Providers, Google, Gemini Search	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Direct Gemini Runtime	9	Google, Model Providers, Faq Models, Testing Live	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Media, Search, and Realtime	10	Google, Google	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Prompt Caching	5	Prompt Caching, Google, Model Providers, Token Use	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No

OpenRouter provider path

Level: M3 Beta
Rationale: Unified provider path is documented and valuable, but model-specific behavior varies.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Provider Setup and Auth	14	Openrouter, Model Providers, Configure, Authentication, Environment, Models, Models	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Chat Runtime and Normalization	15	Openrouter, Model Providers, Prompt Caching	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Provider Recovery and Diagnostics	5	Model Failover, Openrouter, Models	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No
Media Generation and Speech	7	Openrouter, Image Generation, Music Generation, Media Overview, Video Generation, Tts	`Experimental (0%)`	`Alpha (66%)`	`Beta (78%)`	No

Local model providers: Ollama, vLLM, SGLang, LM Studio

Level: M2 Alpha
Rationale: Useful and documented, but environment variance is high.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Provider Setup, Lifecycle, and Diagnostics	12	Local Models, Lmstudio, Ollama, Vllm, Local Model Services, Config Agents, Troubleshooting, Doctor	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Native Provider Plugins	10	Ollama, Lmstudio	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
OpenAI-Compatible Runtime Compatibility	8	Vllm, Sglang, Local Models, Lmstudio	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Local Memory and Embeddings	5	Memory, Doctor	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Network Safety and Prompt Controls	2	Index, Config Tools, Local Models	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No

Long-tail hosted providers

Level: M2 Alpha
Rationale: Many docs/reference pages exist; score should be generated from provider metadata plus live smoke coverage.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Hosted LLM Providers	12	Index, Model Providers, Testing Live, Onboard	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Hosted Media Providers	8	Manifest, Testing Live, Index	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Provider Operations	12	Index, Model Providers, Manifest, Testing Live, Models	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No

Web search tools

Level: M3 Beta
Rationale: Multiple providers and docs exist. Needs quota/error/SSRF proof per provider family.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Search Providers	19	Web, Brave Search, Tavily, Exa Search, Firecrawl, Perplexity Search, Duckduckgo Search, Searxng Search, Gemini Search, Grok Search, Kimi Search, Minimax Search, Ollama Search, Sdk Subpaths, Sdk Overview, Manifest	`Experimental (11%)`	`Beta (79%)`	`Beta (79%)`	No
Setup and Diagnostics	9	Web, Web Fetch, Faq, Api Usage Costs, Brave Search, Perplexity Search, Tavily, Firecrawl	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Network Safety	4	Web, Web Fetch, Firecrawl, Searxng Search	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	No
Tool Availability and Fetch	11	Config Tools, Web Fetch, Web, Faq	`Experimental (18%)`	`Beta (79%)`	`Beta (79%)`	No

Browser automation, exec, and sandbox tools

Level: M3 Beta
Rationale: Core tools are documented, but host security and permission UX should stay under active scorecard review.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Browser Automation	8	Browser Control, Testing, Browser, Index, Audit Checks	`Experimental (13%)`	`Beta (79%)`	`Beta (79%)`	No
Tool Invocation and Execution	6	Exec, Background Process, Tools Invoke Http Api, Operator Scopes, Protocol, Exec Approvals, Exec Approvals Advanced, Elevated	`Experimental (33%)`	`Beta (79%)`	`Beta (79%)`	Yes
Sandbox and Tool Policy	6	Sandboxing, Sandbox Vs Tool Policy Vs Elevated, Multi Agent Sandbox Tools, Codex Harness Reference, Config Tools	`Experimental (0%)`	`Alpha (68%)`	`Beta (79%)`	Yes

Image, video, and music generation tools

Level: M2 Alpha
Rationale: Capability exists across providers, but quality, latency, and parameter compatibility vary too much for beta without per-provider proof.

Area	Capabilities	Docs	Coverage	Quality	Completeness	Long-term support
Media Routing and Discovery	4	Config Agents, Image Generation, Video Generation, Music Generation	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Task Lifecycle and Delivery	12	Media Overview, Image Generation, Video Generation, Music Generation	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Image Generation	9	Image Generation, Infer, Media Overview	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Video Generation	11	Video Generation, Runway, Pixverse, Fal, Openrouter	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No
Music Generation	6	Music Generation	`Experimental (0%)`	`Alpha (61%)`	`Alpha (68%)`	No

Was this useful?

On this page

On this page