Documentation Overview
AxiNode Platform Overview
AxiNode is a zero-credential control plane: outbound-only agents, signed command envelopes, and unified audit. This page orients you to the core model and points you to deeper guides.
Zero credentials
No SSH/RDP secrets are stored. Trust is bound to agent identity + signed envelopes; inbound ports stay closed.
Outbound agents
Lightweight agents dial out with mTLS, verify signatures, execute with least privilege, and stream telemetry.
Unified audit
Every command links user → tenant → agent → output with correlation IDs for forensics and compliance.
How it works (high level)
- Install the agent (Windows/Linux); it generates keys and registers via a short-lived bootstrap token.
- Agent opens outbound mTLS to the control plane; no inbound firewall rules.
- Users/CLI authenticate; commands are signed into envelopes with policy/TTL/nonce.
- Agents verify signature + target binding, then execute in least-privilege sandbox.
- Results, telemetry, and audit entries stream back with correlation IDs.
Where to go next
Detailed docs are split by topic. Start with the guide that matches your task:
- How-to Guide — end-to-end usage (install agent, sign commands, observe audit/telemetry).
- Security — zero-credential model, signed envelopes, outbound-only trust chain.
- Brand & Identity — tone, visuals, and communication rules.
- Purpose & Mission — why we exist, problems we solve (credential theft → lateral movement → domain takeover).
- API Documentation — REST `/api/v1`, gRPC agent channel (`axinode.agent.v1`), ProblemDetails error model.
Quick facts
- Zero-credential: no SSH/RDP secrets stored; trust = agent identity + signatures.
- Outbound-only: no inbound ports, VPNs, or bastions.
- Signed envelopes: TTL, nonce, tenant/server/agent binding, policy enforcement.
- Unified surface: UI, CLI, REST, and gRPC share the same command/audit model.