Skip to main content

Your agents’ credentials, whispered — not shouted

ID Wispera is a credential governance platform that uses the passport/visa metaphor to manage, secure, and audit AI agent credentials. Every credential is encrypted at rest with AES-256-GCM, never stored in plaintext.

Quick start

Get from zero to a working vault with imported credentials in under 3 minutes.

Core concepts

Understand the foundation of ID Wispera’s credential governance model.

Passport model

Learn how passports wrap credentials with identity, authorization, lifecycle, and delegation tracking.

Visa types

Five standard visa classifications plus custom types for credential authorization levels.

Feature guides

Scan & import

Detect exposed credentials across your codebase and import them as governed passports.

Policy engine

Define Cedar-inspired declarative rules to enforce governance policies.

Delegation

Track credential provenance through delegation chains from human owners to AI agents.

Secure sharing

Zero-knowledge credential sharing with AES-256-GCM encryption.

Credential injection

Run subprocesses with vault credentials injected as environment variables via idw exec.

Provisioning

Programmatically create API keys at vendor APIs and wrap them as governed passports.

Locations registry

Auto-detect credential storage paths across 6 built-in providers.

Authentication

Keychain storage, session tokens, IDW_PASSPHRASE, and admin passport management.

Integrations & security

Framework integrations

OpenAI Agents, Google A2A, Slack, LangChain — governed credentials for AI frameworks.

MCP integration

Connect ID Wispera to AI agents like Claude via the Model Context Protocol.

Security architecture

Encryption model, threat model, and security best practices.

Audit log

Complete audit trail, compliance support, and incident response procedures.

SDK reference

Programmatic SDK usage across TypeScript, Python, and Go.

Testing

1,290+ tests across all packages — run and extend the test suite.