Clawd Control
github.com/temaki-ai/clawd-control- Category
- AI Agents
- Rank
- No. 1176Tools index
- Pricing
- Open Source
- Type
- APP
- Use case
- Deployment & Operations
- Interfaces
- Web · CLI
- Builder
- @Temaki-AI
- GitHub
- 115 stars
- Date
About
Real-time dashboard for monitoring and managing Clawdbot AI agents. Provides fleet overview, live metrics, agent creation wizard, and auto-discovery of local agents.
What it does
Clawd Control sits between a browser and one or more Clawdbot gateways. It polls gateway data over WebSockets, separates shared results by agent, and pushes changes to the browser through server-sent events. Operators can inspect sessions, usage, channels, scheduled jobs, machine health, and perform selected agent actions.
Why it's ranked here
The project offers unusually broad Clawdbot operations in a small, understandable package. Monitoring, provisioning, health checks, history collection, authentication, and security audits live in one repository. That breadth is useful, but several operations depend directly on local files, command-line tools, and host conventions, so deployment demands care.
What's good
The collector reuses one WebSocket connection for agents sharing a gateway, then filters sessions and health per agent. Remote gateways can be configured explicitly. Agent creation scaffolds workspaces, registers identities, optionally verifies Telegram credentials, and preserves existing files. Password hashing, session expiry, login throttling, request limits, and audit logging provide practical safeguards.
Tradeoffs
This is specifically for existing Clawdbot installations, not a general agent platform. Sessions remain in memory and disappear when the server restarts. Some metrics assume a Linux proc filesystem and GNU-style disk commands. Several controls invoke local Clawdbot commands or edit agent state files directly. Usage and channel data may remain gateway-wide when agents share a gateway.
How to use it well
Use it on a trusted operations host when you already run several reachable Clawdbot agents and want one browser console for routine checks and interventions. Keep gateway tokens outside configuration through the supported environment override. Treat provisioning and session controls as privileged operations. It does not replace the workspace where humans and agents collaborate.
Technical notes+
package.json defines an ES module package, a clawd-control executable, npm start, and one runtime dependency, ws. server.mjs implements the HTTP server, REST endpoints, SSE broadcasting, password sessions, login rate limiting, agent actions, and security audit integration. collector.mjs exports AgentCollector, groups agents by gateway credentials, speaks protocol version 3 over WebSocket, polls health, sessions, usage, heartbeat, channels, and cron data, then emits per-agent state. discover.mjs reads local Clawdbot configuration and can fall back to scanning agent directories. create-agent.mjs scaffolds workspace content, calls the Clawdbot CLI, patches gateway configuration, optionally verifies and binds Telegram, and signals a hot reload. check.mjs produces pass, fail, and warning reports, while collect-history.mjs stores daily report snapshots. layout.js supplies shared browser state, navigation, themes, responsive CSS, reduced-motion handling, and the SSE client surface.
Observed
- License
- MIT
- Primary language
- JavaScript, using Node.js ES modules plus vanilla browser JavaScript and HTML
- Install surface
- Clone the repository, run npm install, then npm start; package metadata also exposes a clawd-control executable
- Interfaces
- Browser dashboard, REST API, server-sent events, WebSocket gateway client, and command-line self-check and history scripts
- Runtime requirements
- Node.js 18 or newer and reachable Clawdbot agents
- Runtime dependency
- One declared dependency: ws
Read from README.md, package.json, server.mjs, check.mjs, layout.js, discover.mjs, collector.mjs, lucide.min.js, create-agent.mjs, collect-history.mjs, security-lib/auth.js, security-lib/utils.js, security-lib/audit-log.js, security-lib/rate-limit.js.
What it can do
Monitor Clawdbot AI agent fleet in real-time
Connected Clawdbot AI agents → Live metrics and status dashboard
Auto-discover local Clawdbot agents
Local network or system scan → List of detected agents and their details
Create new Clawdbot AI agents
Agent configuration parameters via wizard → Deployed Clawdbot AI agent
Display fleet overview
All registered Clawdbot agents → Consolidated view of agent statuses and metrics
Manage existing Clawdbot agents
Agent selection and management commands → Updated agent configuration or status
Tags
Tech Stack
Comments (0)
No comments yet
Editorially curated, with community endorsements as a secondary signal. Corrections welcome.