Per-user context
Every user gets a persistent identity. The agent tracks history and preferences — no one has to repeat themselves.
SquadClaw · private beta
The AI agent your
whole team talks to.
SquadClaw runs across Matrix, Slack, and Telegram — with persistent memory, tool calling, and scheduled outreach. One deployment. Every channel.
CAPABILITIES
Not a chatbot. An agent.
SquadClaw runs continuously — tracking context per user, executing tools, and proactively reaching out. It acts; you guide.
Memory across sessions
Per-user and per-group memory. Past conversations shape future answers — automatically, across every channel.
Proactive outreach
Cron-driven messages and event-triggered alerts. The agent acts without being prompted.
Full tool ecosystem
Full Model Context Protocol client. Connect any MCP server in minutes — databases, APIs, file systems, custom tools.
One agent, everywhere
Matrix, Slack, Telegram, webhooks. One agent loop, many channels — no duplicate deployments.
Admin control panel
Manage users, groups, tools, and bot identity from a clean web UI. No config files to edit.
CHANNELS
Where your team already talks.
One deployment. Every channel. No context lost between them.
Matrix
End-to-end encrypted. Self-hosted or federated. Threads, media, invite auto-join.
Slack
Socket Mode. DMs, mentions, audio transcription, vision. Works in any workspace.
Telegram
Bot API webhook. DMs, groups, media. Global reach — no extra infra.
SECURITY
Production-grade from day one.
Every access path verified. No soft defaults.
OK
IDOR-safe ownership
Resource access verified against caller identity on every call.
OK
Per-flow rate limiting
Independent limits per endpoint, user, and channel. No shared buckets.
OK
Trusted-proxy XFF
X-Forwarded-For trusted only from configured CIDR ranges.
OK
Invite-accept CAS
Compare-and-swap prevents double-accept race conditions.
Simple, transparent pricing
Start with your own API key. Let us handle the AI when you scale.
BYOK
Bring Your Own Key — connect your OpenAI/Anthropic/DeepSeek API key.
Managed AI
Managed AI — we handle model hosting, caching, and costs.
Enterprise
Contact us Custom volume, SLA, SSO, dedicated infrastructure