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
IntegrationsClaude Desktop

MCP / Claude Desktop

The MCP integration lets Claude Desktop and Cursor call any authenticated API through AgentSecrets. Claude passes a key name. It never sees the credential value.

agentsecrets mcp install

This detects your Claude Desktop or Cursor installation, writes the MCP config automatically, and confirms what was written.

✓ Detected Claude Desktop at ~/Library/Application Support/Claude ✓ Written to claude_desktop_config.json ✓ Restart Claude Desktop to activate

Restart Claude Desktop after running this.

What gets written

{ "mcpServers": { "agentsecrets": { "command": "/usr/local/bin/agentsecrets", "args": ["mcp", "serve"] } } }

No env block. No credential values. The binary path is the only thing in the config.

Manual config

If you prefer to edit claude_desktop_config.json yourself:

{ "mcpServers": { "agentsecrets": { "command": "/usr/local/bin/agentsecrets", "args": ["mcp", "serve"] } } }

Find the config file at:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

Available MCP Tools

AgentSecrets exposes a robust tool surface to your AI assistants, partitioned into logical operational boundaries:

Credential Operations

  • api_call: Executes an HTTP request with credential injection. Your assistant passes key references; the proxy resolves and injects key values from the OS Keychain at the transport boundary.
  • list_keys: Lists available key names in the active project and environment. Returns metadata only, never actual values.
  • check_key: Checks if a specific key name exists in the active project (optionally checking specific environments).
  • get_coverage: Analyzes key parity across environments (development, staging, production) to identify missing configurations.

Context & Environment Management

  • get_status: Returns current session expiry, workspace details, active project configuration, background proxy status, keychain-auth configuration, and cached secrets synchronization state.
  • get_environment: Gets the active environment and details on how it was resolved (env var, project config, or global setting).
  • switch_environment: Sets the active environment (requires confirmation in chat).
  • pull_secrets: Triggers a pull to sync local keyring/cache with the cloud backend (requires confirmation in chat).
  • diff_secrets: Compares local keyring cached keys against cloud secrets.
  • diff_environments: Compares key drift between two specified environments (e.g. source vs. target).

Auditing & Governance

  • get_proxy_logs: Queries the local proxy audit database to view recent requests (method, target domain, key name used, status, and duration).
  • get_blocked_requests: Returns blocked outgoing requests (e.g. domains blocked by allowlist).
  • get_redaction_events: Lists requests where credential leakage was detected and values were redacted.
  • get_audit_summary: Retrieves aggregated statistics on proxy usage.

Access Control & Policy

  • get_allowlist: Returns the list of authorized proxy domains.
  • check_domain: Verifies if a domain is allowed to receive credentials, and returns the CLI command to add it if blocked.
  • get_agent_identity: Checks the active agent's configuration and whether an issued token (agt_...) is active.
  • list_agent_tokens: Lists metadata for all issued agent tokens in the workspace.

Key Lifecycle

  • rotate_key: Permanently deletes/rotates a key from the current environment (requires user confirmation).

To preserve the zero-knowledge model, there is noget_secret tool. The agent can never retrieve raw key values. Credential names are listed, and the proxy performs injections. Outgoing response bodies are automatically scanned and redacted if secret values are leaked in API responses.

Using AgentSecrets with Claude

Once the MCP server is active, Claude can make authenticated API calls by name:

You: Check my Stripe balance Claude: [calls api_call with bearer="STRIPE_KEY", url="https://api.stripe.com/v1/balance"] Claude: Your current balance is $4,200.00 available.

Claude saw the balance. It never saw sk_live_51H....

Was this helpful?
Thanks for your feedback!
Your feedback helps us improve the platform.