$ loading_
Bridge DSH web RPC into MCP tools for workspace and session management.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "dsh-web-mcp" yet — see the docs or source repo.
Use dsh-web-mcp to create a new DSH session in a workspace named "Requirements Analysis", then send a message: Please summarize this week's action items.
Returns the created workspace or session result and the response to the sent message.
Use dsh-web-mcp to resume my previous DSH session and send: Based on the previous discussion, list the next execution steps.
Resumes the specified session and returns the new conversation response.
Use dsh-web-mcp to fetch statistics for the current DSH session, including any available session status data.
Returns session statistics or a status summary to help track usage.
Developers can create, resume, and manage DSH sessions from an MCP-compatible client without directly using the DSH web UI. This helps centralize conversation workflows in one AI workspace.
When teams need to programmatically manage workspaces and send messages to DSH sessions, this bridge exposes those RPC capabilities as standard MCP tools. It fits automation flows or agent orchestration.
If you need session statistics, this tool can retrieve relevant status data from DSH sessions. It is useful for quickly checking progress across multiple sessions.
It is an MCP bridge that exposes the DeepSeek Harness (DSH) web UI's cordis RPC API as stdio tools. This allows any MCP client to manage workspaces and sessions through it.
Based on the provided description, it supports managing workspaces, creating or resuming sessions, sending messages, and fetching session statistics. For more details, see the source repository.
The available information only states that it works as stdio tools for MCP clients and bridges the DSH web UI RPC API. For exact installation steps, runtime requirements, or authentication details, see the source repository.
Search the web, open pages, and find text via an MCP server.
Manage DSH sessions, projects, files, and research locally.
Wrap DeepSeek into standard MCP tools for AI workflows and automation.
Run Codex models through a persistent MCP server without sharing ChatGPT credentials.
Launch DeepSeek Harness Web UI, delegate coding tasks, and review workspace changes.
Connect MCP clients to WebMCP-enabled Chrome pages for tool discovery and execution.