The Trust Layer for the Agent-to-Agent Economy — Ed25519 identity + cryptographic handshake before any transaction. No blockchain required. v0.1
-
Updated
Aug 14, 2026 - Python
The Trust Layer for the Agent-to-Agent Economy — Ed25519 identity + cryptographic handshake before any transaction. No blockchain required. v0.1
DSH says everything is a plugin — so DSH itself can be one too. Run DeepSeek Harness as a codewhale sub-agent / 把 DeepSeek Harness 作为 codewhale 的子 agent(MCP bridge + 黑板 + headless → ACP)
Adapts to YOUR workflow: one MCP server and your existing AI coding agents share environments and read each other's native sessions, memories, and project docs. Claude Code / Codex / Kimi today, extensible to any agent. No extra GUI. Zero-copy.
Session-to-session messaging for AI coding agents — let any agent @ any other, across vendors and devices. Zero cloud, LAN-native.
An interoperability layer between coding-agent CLIs. Any agent delegates bounded work to any other — opencode, Codex, or Claude Code — and gets a durable result back.
Add a description, image, and links to the agent-interop topic page so that developers can more easily learn about it.
To associate your repository with the agent-interop topic, visit your repo's landing page and select "manage topics."