dsh-im-gateway
Run the following command in DeepSeek Harness:
dsh plugin install masquerator-coder/dsh-im-gateway
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install masquerator-coder/dsh-im-gateway to install the plugin from https://github.com/masquerator-coder/dsh-im-gateway into DeepSeek Harness.
About this plugin
If you find yourself ping-ponging between WeChat, QQ, email, and Feishu while wishing you could just ask an AI agent for help without opening a browser, dsh-im-gateway is built for that moment. It bridges six external IM channels — WeChat ilink bot, QQ official bot, SMTP/IMAP email, China Mobile 5G messaging, Feishu official SDK, and a generic HTTP callback — into DeepSeek Harness, routing every incoming message into a persistent Agent session that carries context across turns while keeping separate chats and channels fully isolated.
The plugin is designed around three stages: inbound reception, agent bridging, and outbound delivery. Replies are claimed from the global session-event stream by rpcId so concurrent turns never overwrite each other; sessions survive process restarts without re-creation; and each channel carries its own working directory and model routing so two channels of the same type can never share an agent even if their external chat IDs collide. Built-in safeguards include sender allowlisting, five-second inbound de-duplication, per-session serialization, IM-side approval and user-question push-back, and honest connection-state reporting so a channel that says connected actually answers.
It is aimed at operators, teams, and solo developers who already run agents on DeepSeek Harness and want to extend the conversational surface into the IM apps they use every day — whether that means an email auto-triage assistant, a 5G-message knowledge base query, or a Feishu bot that triggers long-running inference.
Use Cases
- Chat with an AI agent directly inside WeChat, QQ, or Feishu without a browser tab
- Trigger long-running inference via email and receive replies in the same thread
- Unify internal ticket and query workflows through 5G messaging or HTTP webhooks
Best For
- Developers already running agents on DeepSeek Harness
- Product teams embedding AI into multi-entry IM touchpoints
- Solo developers seeking seamless IM-to-agent integration without context switching
Related Plugins
A unified suite combining hot runtime injection, task-aware thinking-mode routing, and a graded session protocol with red-team gates to sustain model diligence across long-horizon inference.
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.