dsh-memos-bridge

A DeepSeek Harness bundle that bridges the MemOS memory service into the agent over MCP. Install the bundle, run one setup script, restart the Harness, and the agent gains persistent-memory tools named mcp__memos__*.

What you get

With the bundle active, the agent can call (a subset of the MemOS MCP surface):

How it works

DeepSeek Harness (web profile)
  └─ cordis.patch.yml  ──inserts──►  @deepseek-ai/dsh-mcp-client  (ships with the dsh CLI)
                                        │  stdio
                                        ▼
                                python -m memos.api.mcp_serve   (MemOS venv)
                                        │
                                        ▼
                        MemOS MOS core: Neo4j (graph memory), Qdrant,
                        LLM + embedding gateway (e.g. Bailian-compatible)

The bundle contributes only a configuration layer (dsh.bundle + cordis.patch.yml); it mounts the stock @deepseek-ai/dsh-mcp-client plugin with a stdio server row. No Harness code is modified.

Prerequisites

• dsh CLI installed (the bundle relies on its built-in @deepseek-ai/dsh-mcp-client).
• A MemOS checkout with its docker stack up (the compose in MemOS/docker provides Neo4j + Qdrant + the MemOS API).
• Python ≥ 3.10 for the MemOS virtualenv.
• The MemOS LLM and embedding gateway reachable from the machine that runs the MCP child (see [Host-run endpoint override](#host-run-endpoint-override)).

Quick start

1. Set up the MemOS side (venv + dependencies + source patches + local tokenizer):

# from the plugin checkout
.\setup.ps1 --memos C:\path\to\MemOS

On POSIX: ./setup.sh --memos /path/to/MemOS. This creates MemOS/.venv, installs MemoryOS[tree-mem] plus python-dotenv, tqdm, langchain_text_splitters, chonkie, applies the required source patches (see below), and downloads a local gpt2 tokenizer.json (HuggingFace mirror first).

2. Install the bundle into a profile:

dsh plugin --profile web add ./dsh-memos-bridge

3. Configure paths (the patch reads these at boot; all optional):

# PowerShell: setx MEMOS_PYTHON "C:\path\to\MemOS\.venv\Scripts\python.exe"
#             setx MEMOS_HOME   "C:\path\to\MemOS"
export MEMOS_PYTHON=/path/to/MemOS/.venv/bin/python
export MEMOS_HOME=/path/to/MemOS

When MEMOS_PYTHON is unset the row falls back to python on PATH; when MEMOS_HOME is unset the child inherits the Harness cwd (MemOS still reads its .env, so point MEMOS_HOME at the checkout unless MemOS is the launch directory).

4. Verify and restart:

dsh --profile web --dump-config    # expect an `id: memos-mcp` row
dsh --profile web                  # restart the GUI; tools appear as mcp__memos__*

Run the smoke test any time:

python scripts/smoke_test.py --python C:\path\to\MemOS\.venv\Scripts\python.exe --memos C:\path\to\MemOS --search

Configuration

The bundle's patch inserts one @deepseek-ai/dsh-mcp-client row (id: memos-mcp, serverName: memos). Environment knobs read by !!js at mount time:

To change any other field (e.g. toolCallTimeoutMs, failOnStartupError), override the row by id in your profile's cordis.patch.yml — later layers win, but an id-targeted patch replaces the whole config, so restate every key:

- id: memos-mcp
  config:
    transport: stdio
    serverName: memos
    command: 'C:/path/to/MemOS/.venv/Scripts/python.exe'
    args: ['-m', 'memos.api.mcp_serve']
    cwd: 'C:/path/to/MemOS'
    toolCallTimeoutMs: 60000
    failOnStartupError: false

Host-run endpoint override

MemOS's .env commonly targets host.docker.internal:18181/18182 (valid inside the MemOS docker network). When the MCP child runs on the host, those endpoints must be reachable from the host. If your gateway is published on the host loopback (or reachable only through a local proxy rule), override the endpoints with an env block on the row:

- id: memos-mcp
  config:
    transport: stdio
    serverName: memos
    command: 'C:/path/to/MemOS/.venv/Scripts/python.exe'
    args: ['-m', 'memos.api.mcp_serve']
    cwd: 'C:/path/to/MemOS'
    env:
      OPENAI_API_BASE: 'http://127.0.0.1:18181/v1'
      MOS_EMBEDDER_API_BASE: 'http://127.0.0.1:18182/compatible-mode/v1'
      MEMRADER_API_BASE: 'http://127.0.0.1:18181/v1'
      QWEN_API_BASE: 'http://127.0.0.1:18181/v1'
    failOnStartupError: false

(These values only stick because the patch script changes MemOS's load_dotenv(override=True) to override=False — ambient env then wins over .env.)

MemOS source patches

scripts/patch_memos.py applies five small, idempotent fixes to the MemOS checkout (tested against MemoryOS 2.0.30):

1. src/memos/api/config.py — load_dotenv(override=True) → load_dotenv(), so host-run env overrides are not clobbered. 2. src/memos/log.py — console handler to stderr; stdout is the MCP protocol channel and log lines there corrupt the stdio stream. 3. src/memos/api/mcp_serve.py — map EMBEDDING_DIMENSION into the default config so the Neo4j vector index matches the embedder dimension. 4. src/memos/mem_os/utils/default_config.py: - env-aware embedder construction honoring MOS_EMBEDDER_BACKEND / MOS_EMBEDDER_API_BASE / MOS_EMBEDDER_API_KEY / MOS_EMBEDDER_MODEL / EMBEDDING_DIMENSION (mirrors APIConfig.get_embedder_config; the MCP default path otherwise ignores them and reuses the chat endpoint); - the sentence-chunker tokenizer points at a local gpt2 tokenizer.json — chonkie otherwise downloads gpt2 from huggingface.co, which is unreachable in some networks.

Run python scripts/patch_memos.py --list to see the patch list. If a patch fails with "not in pre-patch state", your MemOS version differs from 2.0.30 — check the diff and re-apply by hand.

Troubleshooting

Security

The MCP server command runs as trusted executable code outside the agent sandbox (this is why the Harness enables no MCP server by default). Only connect to MemOS servers you run yourself.

License

MIT