985
total weekly downloads
18 packages
n8n-nodes-trigguard
n8n community node: authorize workflow steps via TrigGuard execution gateway (POST /execute)
trigguard
TrigGuard public SDK — authorize, verify, receipts, agents, CLI. One install.
@trigguard/agent-sdk
Canonical TrigGuard agent SDK — authorize PERMIT/DENY/SILENCE/ESCALATE via POST /v1/authorize. Wraps @trigguard/execution-sdk.
@trigguard/cli
TrigGuard CLI — session foundation (tg), execution authority (trigguard)
@trigguard/decision
TrigGuard HTTP decision client: authorize() → POST /decide (no policy in-process)
trigguard-execution-proxy
Execution proxy: policy via POST /execute then forward to HTTP/webhook/Stripe/database connectors
@trigguard/execution-sdk
Execution gateway client: POST /execute, local receipt verification via /.well-known/trigguard/keys.json
@trigguard/express-middleware
Express middleware: TrigGuard POST /execute before route handlers (PERMIT required)
@trigguard/gateway-client
TrigGuard Gateway Client
@trigguard/intent-sdk
Intent-oriented API: maps developer intents to execution surfaces; uses @trigguard/execution-sdk
trigguard-js
TrigGuard JS client: gateway helpers, autoguard, and Express middleware for canonical POST /decide
@trigguard/mcp
TrigGuard MCP — execution authorization over stdio (PERMIT/DENY/SILENCE/ESCALATE)
@trigguard/mcp-server
MCP stdio server for TrigGuard execution authorization (PERMIT/DENY/SILENCE/ESCALATE)
@trigguard/protocol
Shared TypeScript types and JSON contract snapshots for TrigGuard authorization decisions and records (PERMIT|DENY|ESCALATE|SILENCE).
@trigguard/provider-sdk
Public contract for TrigGuard providers: map vendor payloads to ExecutionRequest (no HTTP, no policy)
@trigguard/proxy
Unified execution proxy layer: HTTP proxy app, Express middleware, LangChain pre-auth, execution-sdk helpers
@trigguard/runtime
Opt-in global hooks (e.g. fetch) for automatic authorization via @trigguard/intent-sdk
@trigguard/surface-registry
Execution surface registry: protocol JSON, metadata, hash, validation