Production infrastructure for deploying and running AI agents on Internet Computer canisters

AgentVault Muse Bridge — daemon bridging a Muse agent into E2E-encrypted AgentVault direct messages and rooms.

AgentVault Grok Bridge — daemon bridging a Grok agent into E2E-encrypted AgentVault direct messages and rooms.

AgentVault Claude Bridge — daemon for bridging a Claude agent into secure E2E-encrypted AgentVault 1:1 direct messages and rooms.

The official **OpenClaw** plugin for AgentVault. Adds cryptographic identity, earned trust, and Signal-grade encrypted communications to OpenClaw agents. (Looking for a different framework? See **[Which Framework?](#which-framework)** below.)

Shared cryptographic primitives and telemetry types for the [AgentVault](https://agentvault.chat) ecosystem. Used by `@agentvault/agentvault`, `@agentvault/client`, and the Expo web app.

Lightweight AgentVault client for third-party agents — API key auth + E2E encryption

MCP server for AgentVault — AI-native credential vault for autonomous agents.

Standalone MCP server for AgentVault — exposes encrypted messaging, skill discovery, policy checks, and audit logging as MCP tools

End-to-end encrypted communication channel for AI agents on the [AgentVault](https://agentvault.chat) platform. Connect your agent to its owner with XChaCha20-Poly1305 encryption and Double Ratchet forward secrecy.

AgentVault SDK — the recommended way to integrate agents with the AgentVault secure enclave

Agent Vault: zero-knowledge encrypted backup, restore, and migration for AI agents (OpenClaw, Hermes) on Singularity Layer.

The TrustProvider interface for AgentVault — plus SimpleTrustProvider as a reference implementation. Write your own provider against this contract.

Lightweight SDK for verifying AgentVault agent credentials, trust tiers, and DIDs

TypeScript SDK for Helm (AgentVault) — programmable agent multisig on Solana. Build agents that spend within on-chain policy.

MCP server for Helm (AgentVault) — expose Solana agent-vault operations as Model Context Protocol tools. Read vault/policy state and (with a keypair) propose, approve, and execute guardian-gated transactions.

Shared TypeScript types for AgentVault — the public contract for protocol handlers, credentials, escalations, and webhook events.

A first-party SecureComs agent. It attaches to the local SecureComs daemon and runs a BYO (bring-your-own) OpenAI-compatible LLM against incoming messages, with skills, per-channel memory, and a local tool gate — no AgentVault backend involved.

Agent Vault core: discovery, packaging, and zero-knowledge envelope encryption for AI agent backups (Singularity Layer).

TypeScript SDK for AgentVault — the open-source trust and control layer for AI-agent payments.