Files
netbird/proxy/internal/middleware/builtin/llm_request_parser/factory.go
Maycon Santos b416063bcc [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.
2026-06-27 13:41:00 +02:00

72 lines
2.9 KiB
Go

package llm_request_parser
import (
"bytes"
"encoding/json"
"fmt"
"github.com/netbirdio/netbird/proxy/internal/middleware"
"github.com/netbirdio/netbird/proxy/internal/middleware/builtin"
)
// config is the on-wire config envelope for the middleware.
//
// ProviderID, when set, names the parser to use directly (matched
// against llm.ParserByName, e.g. "openai", "anthropic"). The
// agent-network synthesiser stamps this so requests routed through a
// synthesised provider service don't depend on URL-shape sniffing,
// which is the only signal the middleware otherwise has.
type config struct {
ProviderID string `json:"provider_id,omitempty"`
// RedactPii, when true, runs PII redaction over the captured raw prompt
// before it is emitted as llm.request_prompt_raw — so the
// agent-network access-log row does NOT carry raw emails / SSNs /
// phone numbers even though the framework's per-key redactor (Scan)
// doesn't cover those prompt-shaped patterns. Sourced by the
// synthesiser from the account's redact_pii toggle.
RedactPii bool `json:"redact_pii,omitempty"`
// CapturePrompt gates emission of llm.request_prompt_raw. A nil pointer
// preserves the legacy default (emit), so callers that don't know about
// the toggle (or pre-existing tests with empty config) keep working.
// The synthesiser sets this explicitly to the account's
// enable_prompt_collection toggle: false here suppresses the key
// entirely so the access-log row carries no prompt content at all,
// independent of redact_pii (which only controls the form of the
// content when it IS emitted).
CapturePrompt *bool `json:"capture_prompt,omitempty"`
}
// Factory builds llm_request_parser instances from raw config bytes.
type Factory struct{}
// ID returns the registry identifier.
func (Factory) ID() string { return ID }
// New constructs a middleware instance. Empty, null, and {} configs are
// accepted; non-empty rawConfig that fails to unmarshal is rejected so
// misconfigurations surface at chain build time.
func (Factory) New(rawConfig []byte) (middleware.Middleware, error) {
var cfg config
if len(bytes.TrimSpace(rawConfig)) > 0 {
// Strict decode: a typo'd field (e.g. "capture_prompts") must fail
// chain build rather than silently fall back to the emit-everything
// default and leak prompts.
dec := json.NewDecoder(bytes.NewReader(rawConfig))
dec.DisallowUnknownFields()
if err := dec.Decode(&cfg); err != nil {
return nil, fmt.Errorf("decode config: %w", err)
}
}
// Default capturePrompt to true (legacy emission) when the field is
// absent so non-agent-network callers and pre-toggle tests keep working.
capturePrompt := true
if cfg.CapturePrompt != nil {
capturePrompt = *cfg.CapturePrompt
}
return middlewareImpl{providerID: cfg.ProviderID, redactPii: cfg.RedactPii, capturePrompt: capturePrompt}, nil
}
func init() {
builtin.Register(Factory{})
}