dsh-codebase-memory-mcp
A DeepSeek Harness plugin that bridges the
local codebase-memory-mcp knowledge-graph MCP server into DSH — reusing the
built-in @deepseek-ai/dsh-mcp-client bridge, so you don't implement any MCP
transport yourself.
The model sees tools named mcp__codebase-memory__* (index_repository,
search_graph, trace_path, query_graph, detect_changes, … — 14 tools in
total).
Install (from the marketplace / npm)
dsh plugin --profile web add dsh-codebase-memory-mcp
Then reload the DSH web host (restart the app, or use the in-app reload).
Install (local development)
dsh plugin --profile web add link:D:\path\to\dsh-codebase-memory-mcp
Configure the server binary
The plugin reads its configuration from the codebase-memory-mcp settings
namespace. The default exePath is codebase-memory-mcp, i.e. it assumes the
server binary is on your PATH. If your binary lives elsewhere (e.g. a
downloaded codebase-memory-mcp.exe), set the absolute path in
Settings → Plugins → Plugin configuration → codebase-memory-mcp (or your
profile's settings document). Saving hot-reloads the bridge — no restart needed.
| Setting | Default | Meaning |
|---|
exePath | codebase-memory-mcp | Command/executable for the MCP server. |
cwd | (exe's directory) | Working directory for the spawned server. |
extraArgs | (empty) | Extra CLI args, space-separated, appended to the command. |
Verify
- After reload, ask the model: "What
mcp__codebase-memory__* tools do you have?"
- You should see
mcp__codebase-memory__index_repository, mcp__codebase-memory__search_graph, etc.
- Check the plugin inventory / logs to confirm
mcp-client(codebase-memory) is active and 14 tools were discovered.
How it works
DSH host
└─ composition loads this plugin (cordis.patch.yml insert)
└─ wrapper plugin apply(ctx):
1) ctx.settings.register('codebase-memory-mcp', schema)
2) read scope.get() → build mcp config → ctx.plugin(mcpApply, cfg) (dynamic)
3) scope.watch(() => { handle.dispose(); re-launch }) ← settings change = hot reload
The bridge itself is the built-in @deepseek-ai/dsh-mcp-client; this package
only contributes the composition entry and a runtime-configurable settings
namespace.
Publish
npm publish (registry verifies repository maps back to the same repo, preventing name squatting).- Open a PR adding one entry to awesome-dsh-plugin — the market picks it up automatically, usually within a day.
Known limitations
- No tailored settings card: the value lives in the
codebase-memory-mcp
settings namespace and is edited via DSH's settings document / generic editor.
A point-and-click card would require replicating DSH's client-bundle build.
- If the host cannot resolve
@deepseek-ai/dsh-mcp-client etc. from this
package, add those packages as dependencies or symlink them into the profile.
License
MIT
