dsh-memory-mcp
by tearslee
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DSH_MEMORY_VECTOR | No | Set to off (or 0/false/no) to fully disable vectors and fall back to lexical-only retrieval. | |
| DSH_MEMORY_PROJECT | No | Current workspace directory. Equivalent to --project. Defaults to process.cwd(). | |
| DSH_MEMORY_EMBED_EXE | No | Path to the llama-server executable. | D:\dev\llamacpp\llama-server.exe |
| DSH_MEMORY_SCOPE_ALL | No | When true, memory_save omits scope and defaults to project instead of global. | false |
| DSH_MEMORY_STORE_DIR | No | Memory storage directory. Equivalent to --store-dir. | D:/dev/.dsh-memory-store |
| DSH_MEMORY_EMBED_PORT | No | Port for the embedding server (avoiding DSH's 8078, allowing coexistence). | 8199 |
| DSH_MEMORY_EMBED_MODEL | No | Path to the gguf model file. | D:\dev\models\Qwen3-Embedding-0.6B-Q8_0.gguf |
| DSH_MEMORY_EMBED_POOLING | No | Pooling method. Inferred from model name by default. Must be confirmed when changing models. | |
| DSH_MEMORY_EMBED_THREADS | No | Number of threads for llama-server. | 6 |
| DSH_MEMORY_EMBED_ENDPOINT | No | Directly specify an existing embedding service (e.g. http://127.0.0.1:8078). When set, no process is ever spawned. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Server capabilities have not been inspected yet.
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
No tools | |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
This server cannot be deployed
Maintenance
ActivityMaintained
ResponsivenessNo issues