What is AgentSecrets?
The Zero-Knowledge Difference
How AgentSecrets Works
Installation
Quick Start
Migrating from .env Files
Migrating from Vault / AWS
Migrating from dotenv-vault
Production Checklist
Credential Exposure
What Zero-Knowledge Means
The Proxy Model
The Three-Layer Model
Environments
Agent Identity
Storage Modes
The No get() Principle
Secret-Level Policies
Cloud Overview & Architecture
The Dual-Engine Model
Cloud Resolver Data Plane
Workload & Agent Tokens
Egress Allowlists & Audit Streams
Cloud REST API Reference
Account (init / login)
Server & Self-Hosting (server)
Docs
Shell Autocompletion
Keychain Auth
Secrets
Environments
Credential Proxy
env Injection
Workspaces & Teams
Projects
Agent Identity
Audit & Governance
Integrations Overview
Claude Desktop
Cursor
OpenClaw
HTTP Proxy (Any)
LangChain (Soon)
CrewAI (Soon)
CI/CD Pipeline
SDK Overview
Python SDK
Python API Reference
Python SDK Manual Testing
JavaScript SDK (Soon)
Ecosystem Overview
Zero-Knowledge MCP Server
Server Overview
5-Layer Architecture
Self-Hosting Guide
Authentication & Keys
Workspaces & Teams
Projects & Scope
Environments
Secrets & Sync Protocol
Agent Identity Resolution
Telemetry & Metrics Engine
Audit Log Sync
API Endpoint Reference
Security Overview
Anti-Impersonation & Process Verification
Encryption Model
Zero-Knowledge Sync
Proxy Security Layers
Threat Model
OWASP Top 10 Mitigation
Security FAQ
Third-Party Audit
Reporting Vulnerabilities
Guides Overview
Building on the SDK
Stripe Integration
OpenAI Integration
Multi-Agent Setup
Onboarding Team
CI/CD Pipeline
Publishing ZK MCP
Rotating Credentials
Auditing Team Activity
Dev to Production
Kubernetes Deployment
Monorepo Setup
Production Proxy Hardening
vs .env Files
vs HashiCorp Vault
vs AWS Secrets Manager
vs dotenv-vault
vs Infisical
When Not to Use
Proxy Not Starting
Proxy Not Resolving
Domain Blocked
Sync Conflicts
MCP Not Connecting
Session Token Errors
Proxy Session Authorization
Keychain Storage & Backends
SSRF & Destination Rules
Installation Issues
Error Codes Reference
Frequently Asked Questions
v3.1.x
v3.0.0
v2.1.0
v2.0.0
v1.4.0
v1.3.x
v1.2.0
v1.1.x
v1.0.x
AgentSecrets CloudCloud Resolver Data Plane

Cloud Resolver Data Plane

The Cloud Resolver is the high-throughput Go data plane daemon powering AgentSecrets Cloud.

It acts as a secure intermediary for outbound HTTP/HTTPS requests from AI agents and backend microservices, substituting placeholder handles with real decrypted credentials directly on the wire inside isolated, non-swappable volatile RAM.


Data Plane Architecture

CLOUD RESOLVER INTERNAL ARCHITECTURE ┌──────────────┐ │ Client SDK │ └──────┬───────┘ │ HTTPS Request with AS_SECRET_KEY ┌──────────────────────────────┐ │ Cloud Resolver (:8765) │ │ or POST /v1/resolve │ └──────────────┬───────────────┘ ┌─────────────────────────────────────┼─────────────────────────────────────┐ ▼ ▼ ▼ [ In-Memory State ] [ Allowlist Radix Trie ] [ Volatile Key Store ] - O(1) SHA-256 Token Map - Sub-microsecond match - memguard.LockedBuffer - Capability Gate - Wildcard domain filter - Zero swap to disk │ │ │ └─────────────────────────────────────┼─────────────────────────────────────┘ ┌───────────────────────────────┐ │ Outbound Wire Injection │ │ (Substitutes Real Token) │ └───────────────┬───────────────┘ │ TLS 1.3 Outbound Call Target API (api.stripe.com) ┌───────────────────────────────┐ │ Streaming Response Redactor │ │ (Scans & Scrubs Echoed Keys) │ └───────────────┬───────────────┘ ┌───────────────────────────────┐ │ Lock-Free Audit Buffer │ │ (Asynchronous Batch Queue) │ └───────────────┬───────────────┘ Sanitized Response to Calling Process

Technical Specifications & Security Properties

1Volatile Memory Safety ()

All decrypted workspace Data Encryption Keys (DEKs) and resolved credential buffers are stored exclusively in volatile memory allocated via memguard and protected by mlock(2) system calls:

  • Zero Disk Swapping: Operating system page swapping to disk is strictly disabled for secret buffers.
  • Process Memory Scrubbing: Sensitive byte slices are zeroed out immediately after request serialization.
  • Signal Handling: Process termination signals (SIGINT, SIGTERM) trigger memguard.Purge(), destroying all in-memory keys instantly.

2High-Throughput / Low-Latency Engine

  • Processing Overhead: Introduces less than 0.5 ms processing latency to outbound HTTP calls.
  • Throughput Capacity: Benchmarked to process over 5,000 requests per second per resolver core.
  • Zero-Allocation Pools: Utilizes sync.Pool for byte buffers and streaming readers to eliminate garbage collection pauses on the hot execution path.

3Multi-Pattern Streaming Redaction

If an upstream API returns an error containing an echoed credential (e.g. 401 Unauthorized: Invalid API key sk_live_...), the streaming redaction scanner intercepts the response body and masks all occurrences with [REDACTED]:

Variant FormatRaw Echoed StringRedacted Output
Plaintext Stringsk_live_99ab12cd...[REDACTED]
Base64 Encodedc2tfbGl2ZV85OWFiMTJjZA==[REDACTED]
Hex Encoded736b5f6c6976655f39396162...[REDACTED]
URL Encodedsk_live%5F99ab12cd...[REDACTED]
JSON Escaped"sk_live_99ab12cd...""[REDACTED]"

Invocation Interfaces

The Cloud Resolver exposes two standard transport interfaces:

Interface 1: Forward HTTPS Proxy (:8765)

Standard HTTP CONNECT tunneling and transparent proxy routing for tools and SDKs that support HTTP_PROXY:

export HTTP_PROXY="https://resolver.agentsecrets-website.vercel.app:8765"

Interface 2: Direct Enveloping REST API (POST /v1/resolve)

For serverless runtimes (AWS Lambda, Cloudflare Workers) and AI agent tool calling definitions where running a proxy daemon is impractical, send structured JSON payloads to POST /v1/resolve.


Health & Diagnostic Probes

The Cloud Resolver exposes a standardized health probe on /healthz:

curl -i https://resolver.agentsecrets-website.vercel.app/healthz

Response (200 OK):

{ "status": "healthy", "service": "agentsecrets-cloud-resolver", "timestamp": "2026-08-23T00:00:00Z" }
Was this helpful?
Thanks for your feedback!
Your feedback helps us improve the platform.