dsh-openclaw-acp
Run the following command in DeepSeek Harness:
dsh plugin install BeAChanger/dsh-openclaw-acp
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install BeAChanger/dsh-openclaw-acp in DeepSeek Harness to install this plugin; source is at https://github.com/BeAChanger/dsh-openclaw-acp .
About this plugin
Teams running DeepSeek Harness for private or local model inference often find it tedious to open a terminal for every prompt. dsh-openclaw-acp solves one specific problem: it exposes Harness profiles to OpenClaw through the official Agent Client Protocol transport, so you can send messages and receive results directly from WeChat or other messaging channels.
The design hinges on a clean separation of concerns. Harness owns the agent, model, tools, workspace sandbox, and session log. OpenClaw ACPX manages ACP process lifecycle, dispatch, and conversation routing. The channel plugin handles the messaging transport. None of these layers duplicate the others, and no user credential crosses the ACP boundary. The default route targets deepseek-v4-flash with thinking enabled, maximum reasoning effort, a one-million-token context window, and a 384,000-token output cap, ready for heavier inference tasks out of the box.
This plugin is a good fit for teams and developers who already operate DeepSeek Harness and OpenClaw and want to shift the inference entry point from a CLI to a messaging channel. Security defaults include workspace-write permission mode, with the Gateway and Harness running under a dedicated OS account, making it suitable for production deployments with strict data-boundary requirements.
Use Cases
- Invoke DeepSeek Harness reasoning tasks directly from WeChat or other messaging channels
- Expose inference capabilities to the OpenClaw ecosystem without duplicating Harness
- Chain model inference, conversation routing, and messaging transport over a single ACP pipeline
Best For
- Teams that already run DeepSeek Harness and need multiple entry points
- Developers managing messaging-channel routing via OpenClaw
- Ops personnel migrating inference entry points from CLI to messaging platforms
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.