Mnemozine
by DiverOfDark
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MNEMOZINE_MCP_HOST | No | MCP bind host | 127.0.0.1 |
| MNEMOZINE_MCP_PORT | No | MCP bind port | 8765 |
| MNEMOZINE_LOG_LEVEL | No | Logging level | INFO |
| MNEMOZINE_FALKORDB__URL | No | FalkorDB (Redis protocol) connection URL | redis://localhost:6379 |
| MNEMOZINE_EMBEDDING__MODEL | No | Ollama embedding model | bge-m3 |
| MNEMOZINE_RETRIEVAL__TOP_K | No | Default results per scoped query | 10 |
| MNEMOZINE_EXTRACTION__MODEL | No | LiteLLM provider/model id | openai/qwen2.5 |
| MNEMOZINE_MAINTENANCE__CRON | No | Scheduled maintenance cadence | 0 3 * * * |
| MNEMOZINE_FALKORDB__PASSWORD | No | Optional FalkorDB/Redis password | |
| MNEMOZINE_EMBEDDING__BASE_URL | No | Ollama base URL | http://localhost:11434 |
| MNEMOZINE_EXTRACTION__API_KEY | No | API key for extraction endpoint | not-needed |
| MNEMOZINE_EMBEDDING__TIMEOUT_S | No | Per-request timeout (seconds) | 60 |
| MNEMOZINE_EXTRACTION__BASE_URL | No | OpenAI-format base URL for extraction LLM | http://localhost:8000/v1 |
| MNEMOZINE_FALKORDB__GRAPH_NAME | No | Graphiti graph/keyspace name | mnemozine |
| MNEMOZINE_INJECT__TOKEN_BUDGET | No | Hard cap for injection token budget; truncate, never overflow | 500 |
| MNEMOZINE_EMBEDDING__DIMENSIONS | No | Vector dimensionality (bge-m3 is 1024-d) | 1024 |
| MNEMOZINE_EXTRACTION__TIMEOUT_S | No | Per-request timeout (seconds) | 120 |
| MNEMOZINE_EXTRACTION__TEMPERATURE | No | Extraction determinism | 0.0 |
| MNEMOZINE_INGEST__CHUNK_MAX_CHARS | No | Maximum characters per episode chunk | 8000 |
| MNEMOZINE_INGEST__STRIP_TOOL_CALLS | No | Strip tool_calls/tool results on ingest | true |
| MNEMOZINE_CROSSREF__MAX_SUGGESTIONS | No | Maximum cross-reference suggestions (1–2) | 2 |
| MNEMOZINE_INGEST__CLAUDE_CONFIG_DIR | No | Root of Claude Code config/transcripts | ~/.claude |
| MNEMOZINE_INGEST__CHUNK_MAX_MESSAGES | No | Maximum messages per episode chunk | 40 |
| MNEMOZINE_INGEST__CLEANUP_PERIOD_DAYS | No | Claude Code's cleanupPeriodDays retention | 30 |
| MNEMOZINE_MAINTENANCE__MAX_NODE_DEGREE | No | Traversal-bound cap | 64 |
| MNEMOZINE_RETRIEVAL__NEIGHBORHOOD_HOPS | No | Entity-neighborhood traversal depth | 1 |
| MNEMOZINE_CROSSREF__RELEVANCE_THRESHOLD | No | Cross-reference relevance threshold (precision over recall) | 0.8 |
| MNEMOZINE_MAINTENANCE__EDGE_WEIGHT_FLOOR | No | Low-weight edge pruning threshold | 0.1 |
| MNEMOZINE_INJECT__MAX_PREFERENCE_SNIPPETS | No | Maximum top-preference snippets in the index | 5 |
| MNEMOZINE_RETRIEVAL__P95_LATENCY_TARGET_MS | No | P95 latency target in milliseconds | 500 |
| MNEMOZINE_MAINTENANCE__DECAY_HALF_LIFE_DAYS | No | Decay half-life in days | 30 |
| MNEMOZINE_CROSSREF__VECTOR_FALLBACK_THRESHOLD | No | Minimum cosine sim for vector fallback in cross-reference | 0.75 |
| MNEMOZINE_MAINTENANCE__DECAY_ARCHIVE_AFTER_DAYS | No | Hot-to-archive demotion delay in days | 90 |
| MNEMOZINE_MAINTENANCE__CONTRADICTION_CANDIDATE_CAP | No | Supersede-LLM candidate cap | 5 |
| MNEMOZINE_MAINTENANCE__DEDUP_EQUIVALENCE_THRESHOLD | No | Dedup equivalence threshold (reinforce-vs-add) | 0.9 |
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 | |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
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/DiverOfDark/Mnemozine'
If you have feedback or need assistance with the MCP directory API, please join our Discord server