[management,proxy] Agent network: per-account LLM gateway (policy, metering, multi-provider) (#6555)

* [agent-network] Shared proto, OpenAPI schema, and generated types

* [agent-network] Management: store, manager, synthesizer, policy engine, provider catalog, HTTP/gRPC API

Adds the account-scoped agent-network module: provider/policy/budget CRUD and
store, the reverse-proxy service synthesizer, policy selection + limit
enforcement, the provider catalog (incl. Vertex AI and AWS Bedrock entries),
and the management HTTP + proxy gRPC surfaces.

* [management] Fix agent-network proxy-peer fan-out on affected-peer recompute

The affected-peers resolver loaded only persisted reverse-proxy services, but
agent-network services are synthesized on demand and never persisted. As a
result the embedded proxy peer was never folded into the affected set when a
client's group changed, so the proxy received no network-map update for a newly
authorised client and rejected its handshake until a full resync (restart).

loadProxyServices now merges the synthesized agent-network services (injected
via a registration hook to avoid an import cycle), so proxy peers learn newly
authorised clients immediately.

* [proxy] Reverse-proxy middleware framework, chain, and request plumbing

The per-target middleware chain (slots, dispatcher, mutation gate, metadata
merger), body capture, access-log terminal sink, and the proxy wiring that
builds + runs chains for synthesized agent-network services.

* [proxy] LLM parsers, pricing, and builtin middlewares (OpenAI, Anthropic, Vertex AI, AWS Bedrock)

Request/response parsers and SSE/event-stream metering, the embedded pricing
table, and the builtin middleware set: request parser, router, policy
limit-check/record, cost meter, guardrail, identity inject, response parser.
Includes the path-routed providers — Google Vertex AI (keyfile:: service-account
OAuth minting) and AWS Bedrock (bearer auth, invoke/converse/streaming, optional
/bedrock prefix) — plus the Models allowlist and unmeterable-publisher deny.

* [proxy] IPv6 in-place apply and TCP accept-loop hardening on netstack listeners

* [agent-network] End-to-end test suite, module docs, and deployment preset

* [agent-network] Fix codespell typos and exclude false positives

- labelgen word pool: vermillion -> vermilion, racoon -> raccoon.
- codespell ignore list: add flate (Go compress/flate package), recordin
  (a test-local identifier), and unparseable (a valid alternative spelling used
  consistently across identifiers + a metadata-value constant).

* [management] Set LastSeen on injected proxy peer in realstack test (MySQL strict-mode)

The injected embedded proxy peer had a PeerStatus with a zero LastSeen, which
serializes to '0000-00-00' and is rejected by MySQL in strict mode (SQLite
tolerates it). Set LastSeen to a valid time so SaveAccount succeeds on both
engines.

* [agent-network] Remove e2e shell-script suite from this branch

The end-to-end shell scripts under scripts/e2e/ are maintained in a separate
testing suite and are not part of this change set.

* [agent-network] Polish module docs: remove internal review scaffolding, fix links, verify diagrams

Strip PR-review framing, commit references, absolute paths, and stale internal
references from the agent-network module docs; fix broken relative links; verify
all diagrams against the current architecture. Remove the internal AI-reviewer
prompt file.

* [management] Refine session expiration handling to support 3-state encoding for SSO deadlines

* [agent-network] Relocate agentnetwork package to internals/modules

Move management/server/agentnetwork (and its catalog/, labelgen/, types/
subpackages) to management/internals/modules/agentnetwork, alongside the
reverse-proxy module, and rewrite all importers. Pure relocation: package names,
the synthesizer + affectedpeers registration hook, and store access (shared
store.Store) are unchanged, so no import cycle is introduced (affectedpeers
still depends only on the agentnetwork/types leaf).

* [agent-network] Co-locate HTTP handlers in the module (RegisterEndpoints)

Move the agent-network HTTP handlers from server/http/handlers/agentnetwork into
the module at internals/modules/agentnetwork/handlers (package handlers) and
rename the entrypoint AddEndpoints -> RegisterEndpoints, matching the
reverse-proxy module convention. Wiring in http/handler.go updated accordingly.
This commit is contained in:
Maycon Santos
2026-06-27 13:41:00 +02:00
committed by GitHub
parent 615631567a
commit b416063bcc
187 changed files with 36835 additions and 660 deletions

View File

@@ -116,6 +116,24 @@ func (c *Controller) OnPeerDisconnected(ctx context.Context, accountID string, p
c.EphemeralPeersManager.OnPeerDisconnected(ctx, peer)
}
// injectAllProxyPolicies prepares an account for the per-peer network-map
// computation. It prepends the in-memory agent-network services synthesised
// from the account's current provider/policy state to account.Services so
// the existing InjectProxyPolicies + injectPrivateServicePolicies walks pick
// them up alongside persisted reverse-proxy services. Synthesised services
// are never persisted; the account is loaded fresh per cycle so re-prepending
// is safe and idempotent. Accounts without agent-network providers get an
// empty synth slice — no behaviour change.
func (c *Controller) injectAllProxyPolicies(ctx context.Context, account *types.Account) {
synth, err := c.repo.SynthesizeAgentNetworkServices(ctx, account.Id)
if err != nil {
log.WithContext(ctx).Warnf("synthesise agent-network services for account %s: %v", account.Id, err)
} else if len(synth) > 0 {
account.Services = append(synth, account.Services...)
}
account.InjectProxyPolicies(ctx)
}
func (c *Controller) CountStreams() int {
return c.peersUpdateManager.CountStreams()
}
@@ -150,7 +168,7 @@ func (c *Controller) sendUpdateAccountPeers(ctx context.Context, accountID strin
var wg sync.WaitGroup
semaphore := make(chan struct{}, 10)
account.InjectProxyPolicies(ctx)
c.injectAllProxyPolicies(ctx, account)
dnsCache := &cache.DNSConfigCache{}
dnsDomain := c.GetDNSDomain(account.Settings)
peersCustomZone := account.GetPeersCustomZone(ctx, dnsDomain)
@@ -281,7 +299,15 @@ func (c *Controller) sendUpdateForAffectedPeers(ctx context.Context, accountID s
var wg sync.WaitGroup
semaphore := make(chan struct{}, 10)
account.InjectProxyPolicies(ctx)
// The affected-peer path MUST mirror sendUpdateAccountPeers (line 171)
// here: injectAllProxyPolicies prepends the synthesised agent-network
// services BEFORE InjectProxyPolicies + private-service policies run.
// Previously this path called only account.InjectProxyPolicies, which
// skipped the synth-services prepend — so peer-level changes
// (proxy restart, embedded peer connect/disconnect) propagated a
// network map that omitted the synth DNS zone, and the agent kept
// resolving against the stale or absent record.
c.injectAllProxyPolicies(ctx, account)
dnsCache := &cache.DNSConfigCache{}
dnsDomain := c.GetDNSDomain(account.Settings)
peersCustomZone := account.GetPeersCustomZone(ctx, dnsDomain)
@@ -399,7 +425,7 @@ func (c *Controller) UpdateAccountPeer(ctx context.Context, accountId string, pe
return fmt.Errorf("failed to get validated peers: %v", err)
}
account.InjectProxyPolicies(ctx)
c.injectAllProxyPolicies(ctx, account)
dnsCache := &cache.DNSConfigCache{}
dnsDomain := c.GetDNSDomain(account.Settings)
peersCustomZone := account.GetPeersCustomZone(ctx, dnsDomain)
@@ -603,7 +629,7 @@ func (c *Controller) GetValidatedPeerWithMap(ctx context.Context, isRequiresAppr
return nil, nil, 0, err
}
account.InjectProxyPolicies(ctx)
c.injectAllProxyPolicies(ctx, account)
approvedPeersMap, err := c.integratedPeerValidator.GetValidatedPeers(ctx, account.Id, maps.Values(account.Groups), maps.Values(account.Peers), account.Settings.Extra)
if err != nil {
@@ -874,7 +900,7 @@ func (c *Controller) GetNetworkMap(ctx context.Context, peerID string) (*types.N
return nil, err
}
account.InjectProxyPolicies(ctx)
c.injectAllProxyPolicies(ctx, account)
resourcePolicies := account.GetResourcePoliciesMap()
routers := account.GetResourceRoutersMap()
groupIDToUserIDs := account.GetActiveGroupUsers()