Implemented in the project
- Local-first Gateway, CLI onboarding, and Control UI
- WebChat plus documented messaging channel adapters
- Workspace skills, multi-agent routing, and companion nodes
CinaClaw / open-source personal AI assistant.
Run an assistant through your own local-first Gateway.
CinaClaw provides an installable CLI, Gateway, Control UI, messaging adapters, skills, and optional companion apps. Channel access and model credentials remain configuration the operator controls.
Product evidence / current scope.
What is verifiable today.
Status and capabilities below are tied to public project repositories. They do not imply a customer deployment, managed SLA, or production availability beyond the linked evidence.
Status: Open-source release
The installable CLI and local-first Gateway document stable, beta, and development channels; external channels require user configuration.
Verified: · commit 4975cf7
Documented interfaces and integrations
- CLI and Gateway WebSocket/RPC
- WhatsApp, Telegram, Slack, Discord, Signal, and other documented adapters
- macOS, Windows, iOS, Android, and Linux companion surfaces
CinaClaw
Example data generated by this site—not a customer screenshot or proof of a live deployment.
Input Inbound support message
Control Context + explicit approval
Output Draft reply / routed action
Evidence sources
Capabilities.
A focused layer for the work your team needs to finish.
These are reference workflow patterns, not a promise that every capability is enabled in every release or deployment. Start with the smallest useful workflow and confirm current availability against the evidence above.
Install through the guided CLI.
Configure the Gateway, workspace, models, channels, and skills from a documented onboarding path.
Keep the control plane local-first.
Run sessions, channels, tools, and events through a Gateway under the operator’s control.
Connect documented messaging adapters.
Use WebChat or configure supported adapters such as Telegram, Slack, Discord, Signal, and others.
Pair unknown direct-message senders.
Keep the default pairing policy for common messaging channels unless the operator explicitly changes it.
Load focused workspace capabilities.
Use bundled, managed, or workspace skills with explicit instructions and supporting resources.
Add optional companion surfaces.
Connect documented Windows, macOS, Linux, iOS, or Android surfaces when the setup requires them.
Workflow.
Move from install to one deliberately configured assistant path.
A reference setup confirms the local Gateway, one model, one channel, and the applicable access policy before expanding.
01 / Run onboarding Install the CLI and configure the Gateway, workspace, model authentication, and daemon mode.
02 / Connect one channel Start with WebChat or one supported adapter and keep its inbound access policy explicit.
03 / Add a focused skill Introduce only the instructions and tools required for the first job.
04 / Verify and operate Check Gateway status, exercise the channel, and use the security runbook before remote exposure.
Illustrative local configuration.
Keep the implementation understandable.
Use explicit inputs, review points, and outputs so the system is easier to operate and change.
assistant "local" {
gateway = "self-hosted"
channels = ["webchat", "configured-adapters"]
dmPolicy = "pairing"
}Operating principles.
Built for practical adoption.
Operator control.
The operator chooses the models, channels, tools, access policies, and exposure path.
Secure defaults.
Pairing, allowlists, sandbox settings, and exposure guidance should be reviewed before use.
Incremental setup.
Validate one model, channel, and skill path before adding more integrations.
Next step.
Map this product to one real workflow.
Tell us the work, systems, constraints, and review points your team already has. We will help identify a sensible starting scope.