dsh-agent-relay
Run the following command in DeepSeek Harness:
dsh plugin install Noelune/dsh-agent-relay
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, run the command: dsh plugin install Noelune/dsh-agent-relay. The full source repository is available at https://github.com/Noelune/dsh-agent-relay.
About this plugin
In today's AI development landscape, many agent frameworks (such as Codex and Claude Code) focus on single-agent reasoning and tool usage, yet they lack standardized communication mechanisms between agents. When multiple independent agents run concurrently on the same machine, they cannot directly engage in code reviews, fact-checking, or collaborative task distribution, creating a bottleneck for teamwork. dsh-agent-relay addresses this architectural gap by providing a fully decoupled, lightweight local communication relay bus, enabling developers to easily build collaborative links for agent fleets.
The plugin's core strengths lie in its secure and reliable design: it defaults to binding to the local loopback address (Loopback-First), eliminating cloud dependencies, while ensuring rigorous authentication through HMAC-SHA256 signature verification. The message routing employs high-reliability delivery with features like a 7-day TTL queue, exponential backoff retries, and idempotent deduplication, safeguarding data integrity and privacy. For DeepSeek Harness users, it offers first-class plugin integration, registering multiple model tools and providing a graphical status panel for seamless and intuitive inter-agent collaboration.
dsh-agent-relay is ideal for developers and researchers running multiple AI agents in local environments, especially teams using DeepSeek Harness or similar tools to build agent fleets. If you need to securely exchange messages, share tasks, or validate results among different agents (like dsh, Codex, or Python clients) without introducing complex external services or compromising data privacy, this plugin serves as an excellent lightweight solution. It allows you to focus on collaborative innovation with agents rather than the intricacies of underlying communication setup.
Use Cases
- Enables secure message exchange and communication when multiple independent AI agents run on the same machine.
- Facilitates direct code review, fact cross-validation, and collaborative task distribution between agents.
- Acts as a lightweight communication bus to build collaborative workflows for local agent fleets.
Best For
- Developers using DeepSeek Harness or local AI agent frameworks.
- Teams or researchers building multi-agent collaborative workflows.
- AI practitioners who prioritize data privacy and local deployment security.
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.