mirror of
https://github.com/netbirdio/netbird.git
synced 2026-07-16 11:39:57 +00:00
Update readme
This commit is contained in:
@@ -1,10 +1,19 @@
|
||||
# NetBird Agent Network
|
||||
|
||||
Agent Network is NetBird's access control layer for AI agents and the people who run
|
||||
them. It gives every agent a real identity, tied to your identity provider (IdP), and
|
||||
governs what it can reach — the LLM APIs and AI gateways it can call, and the internal
|
||||
resources it can access. Traffic flows only over the encrypted NetBird tunnel, scoped by
|
||||
policy, with no API keys to leak.
|
||||
Agent Network is NetBird's access control layer for AI agents and the people who run them.
|
||||
It gives every agent a real identity, tied to an identity provider (IdP), and governs what it can reach: LLM APIs and
|
||||
AI gateways it can call, and the internal resources it can access. Traffic flows only over the encrypted NetBird tunnel,
|
||||
scoped by policy, with no API keys or other credentials to leak.
|
||||
|
||||
It also gives you control over cost. Because every LLM request passes through an
|
||||
identity-aware proxy, you can:
|
||||
|
||||
- **Set spending and rate limits** per agent, per user, or per team — with hard caps
|
||||
that stop requests once a budget is reached.
|
||||
- **Restrict models and providers** so agents can only call approved (and cost-appropriate)
|
||||
endpoints, keeping expensive models off-limits unless explicitly allowed.
|
||||
- **Attribute usage** by tracking token consumption and cost per identity, so every
|
||||
request is tied back to the agent and person responsible.
|
||||
|
||||
> **Beta.** Agent Network is open source and can be self-hosted on your own
|
||||
> infrastructure.
|
||||
|
||||
Reference in New Issue
Block a user