Myelin
by et-do
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MYELIN_DATA_DIR | No | Directory where all data lives. Default: ~/.myelin | |
| MYELIN_LATERAL_K | No | Max results per session/scope (0 = off). Default: 1 | |
| MYELIN_LOG_LEVEL | No | Logging verbosity (structured JSON to stderr). Default: INFO | |
| MYELIN_MULTIPROBE | No | 3-probe retrieval (original + keywords + entity-expanded). Default: true | |
| MYELIN_MAX_MEMORIES | No | Hard memory cap; 0 = unlimited. LRU eviction when exceeded. Default: 0 | |
| MYELIN_CHUNK_MAX_CHARS | No | Max characters per chunk. Default: 1000 | |
| MYELIN_EMBEDDING_MODEL | No | Bi-encoder model. Default: all-MiniLM-L6-v2 | |
| MYELIN_NECORTEX_RERANK | No | Cross-encoder re-ranking. Default: true | |
| MYELIN_NECORTEX_WEIGHT | No | CE/bi-encoder blend (0.0–1.0). Default: 0.6 | |
| MYELIN_DEFAULT_N_RESULTS | No | Results returned per recall. Default: 5 | |
| MYELIN_MIN_CONTENT_LENGTH | No | Minimum chars to pass the input gate. Default: 20 | |
| MYELIN_RECALL_OVER_FACTOR | No | Over-retrieval multiplier for re-ranking headroom. Default: 8 | |
| MYELIN_CHUNK_OVERLAP_CHARS | No | Overlap between text chunks. Default: 200 | |
| MYELIN_CROSS_ENCODER_MODEL | No | Cross-encoder model. Default: ms-marco-MiniLM-L-6-v2 | |
| MYELIN_CONSOLIDATION_INTERVAL | No | Auto-consolidate every N stores (0 = disabled). Default: 50 | |
| MYELIN_DEDUP_SIMILARITY_THRESHOLD | No | Above this = near-duplicate, rejected. Default: 0.95 |
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 | |
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/et-do/myelin'
If you have feedback or need assistance with the MCP directory API, please join our Discord server