Local CPU context preparation and saved task notes through an unofficial DSH MCP integration.
DSH integration
Compatible
Author-claimed
Safety audit
Unaudited
Last verified
2026-10-30
License
MIT
01What can it help you accomplish?
Prepare context and reuse saved task notes with DSH
Budgeted context and recalled decisions, constraints and next steps
Users connecting DeepSeek Harness to TOMC through local MCP
02How to install into DeepSeek Harness
Prerequisites
- Python 3.10 or later, the `.[mcp]` extra, Node.js and a configured DSH model
Installation steps
- 01
Clone the repository and activate a local Python environment.
- 02
Install MCP support with `python -m pip install -e '.[mcp]'`.
- 03
Use the README’s `docs/deepseek_harness.md` link for local MCP installation and the DSH two-session demonstration.
Verify the integration
Not specified by the author
03DSH integration and capability boundaries
Unofficial local MCP integration for DeepSeek Harness; the README links a DSH installation and two-session demonstration guide.
Prepare task context
A history, next task and memory budget→Selected source text and supported state, relation or count records
Successful MCP preparation and recall save the prepared context and task locally by default. Turning off context recording stops future storage; existing snapshots remain until deleted.Save and recall task notes
Explicit `remember_memory` and `recall_memory` calls→Saved briefs, decisions, constraints and unfinished work for the next task
Successful MCP preparation and recall save the prepared context and task locally by default. Turning off context recording stops future storage; existing snapshots remain until deleted.
04Who is it for? When not to use it?
Good for
- Users connecting DeepSeek Harness to TOMC through local MCP
Not for
- The plugin does not automatically record the entire chat.
- Shared notebooks reuse saved notes; they do not migrate native chats or provide cross-device cloud sync.
- The memory token estimate excludes the task, chat framing, reader output and inspector.
- Successful MCP preparation and recall save the prepared context and task locally by default. Turning off context recording stops future storage; existing snapshots remain until deleted.
05Compatibility, maintenance and safety notes
- The plugin does not automatically record the entire chat.
- Shared notebooks reuse saved notes; they do not migrate native chats or provide cross-device cloud sync.
- The memory token estimate excludes the task, chat framing, reader output and inspector.
- Successful MCP preparation and recall save the prepared context and task locally by default. Turning off context recording stops future storage; existing snapshots remain until deleted.
Source and README available; claims are author statements.
06Frequently asked questions
How does TOMC connect to DeepSeek Harness?
The README lists an unofficial local MCP integration and links `docs/deepseek_harness.md` for installation and a two-session demonstration.
Does installing TOMC capture every DSH conversation?
No. Saving and recalling task notes are explicit tool calls; the plugin does not automatically record an entire chat.
Can two assistants share TOMC notes?
Yes, when both clients use the same local TOMC database. The default is `~/.tomc/memory.sqlite3`; this is saved-note sharing rather than native chat migration or cloud sync.
07Related DSH workflows
weknora
by tencent
Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
honcho
by plastic-labs
Memory library for building stateful agents
webnovel-writer
by lingfengqaq
A long-form Chinese web-novel writing assistant for Claude Code and DeepSeek Harness, designed to reduce memory loss and hallucinations in serialized projects of up to 2 million characters.
mirage
by strukto-ai
The World's First Virtual Terminal for AI Agents
08Data and sources
| DeepSeek Harness | [Local MCP installation and two-session demo](docs/deepseek_harness.md) | Python with `.[mcp]`, Nod…
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-10-30. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.