dsh-lark-bridge
Run the following command in DeepSeek Harness:
dsh plugin install chenxin105/dsh-lark-bridge
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install chenxin105/dsh-lark-bridge in your DeepSeek Harness terminal to install; the full source lives at https://github.com/chenxin105/dsh-lark-bridge . On first launch the Web UI will prompt you for the Feishu App ID and Secret.
About this plugin
Feishu and Lark ecosystems hold abundant docs, Bitable sheets, and AI agents, yet until now AI coding agents could only rely on users to copy-paste content into the conversation, a slow and error-prone loop. dsh-lark-bridge connects the DSH agent directly to the Feishu/Lark Open Platform so the model can invoke platform APIs on its own, eliminating the manual relay step.
The plugin ships a full outbound tool set: send text or card messages, read docx documents as plain text or structured blocks, list/create/update/batch-create Bitable records, trigger a Feishu bot or agent, and start an Aily skill via server-side API. Authentication resolves tenant_access_token with auto-refresh through the credentials domain, keeping secrets out of every configuration response. With Phase 2 inbound enabled, a Feishu private-chat message can fire a full DSH agent turn and return a reply in the same thread, closing the conversational loop.
Best suited for developers and teams who work inside Feishu or Lark and want the AI agent to pull context, operate on tables, and trigger bot workflows autonomously. If you are tired of pasting links and waiting for the model to guess, this plugin is built for that workflow.
Use Cases
- Let the AI agent pull Feishu docs and Bitable sheets as context without copy-paste
- Trigger a DSH agent from a Feishu private chat and get replies in the same thread
- Batch-create or update Bitable records through agent tool calls instead of manual entry
Best For
- Developers in Feishu or Lark who need AI agents to operate platform data natively
- Teams building Feishu private-chat-driven agent workflows
- Engineers in the DSH ecosystem integrating with the Feishu Open Platform
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.