mnemos
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DATABASE_URL | No | PostgreSQL connection string (e.g., postgresql://user:pass@host:5432/mnemos). Required when using the default postgres backend. | |
| GROQ_API_KEY | No | API key for Groq. Required if LLM_PROVIDER is 'groq'. | |
| LLM_PROVIDER | No | Which LLM provider to use for generation. Must be 'anthropic' or 'groq'. | |
| NEO4J_PASSWORD | No | Password for the Neo4j database. Required when STORAGE_BACKEND is 'neo4j'. | |
| STORAGE_BACKEND | No | Storage backend for episodic and semantic memory. One of 'postgres' (default), 'qdrant', or 'neo4j'. | |
| ANTHROPIC_API_KEY | No | API key for Anthropic. Required if LLM_PROVIDER is 'anthropic'. | |
| MNEMOS_MCP_USER_ID | No | Fixed user ID for MCP memory store. Defaults to 'claude-code'. |
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
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| mnemos_rememberA | Store a durable fact worth recalling in a future session — a stated preference, a decision and its reasoning, a recurring convention. Not routine task details, not anything already in the codebase/CLAUDE.md. |
| mnemos_recallA | Retrieve memories relevant to a query — call this near the start of a task that might benefit from prior context about the user or their preferences/projects. |
| mnemos_list_memoriesA | List all currently active memories with their IDs, for browsing or as input to mnemos_forget. |
| mnemos_forgetA | Forget (archive) a specific memory by its ID, as shown by mnemos_list_memories. Soft-delete: the fact is archived, not erased, and stays visible in mnemos's audit log — never call this speculatively, only when the user explicitly asks to forget something. |
| mnemos_reflectA | Run a consolidation pass: merge near-duplicate memories, decay stale ones, archive anything that decays past the forget threshold. Requires a real LLM provider configured (ANTHROPIC_API_KEY or GROQ_API_KEY) for the merge step. Call this occasionally to keep memory tidy, not on every turn. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool maps to a distinct lifecycle stage: remember stores, recall retrieves relevant memories, list_memories browses all memories, forget archives by ID, and reflect consolidates. There is no meaningful overlap in intended use.
All tools share a consistent mnemos_ prefix and lowercase snake_case imperative style. The only minor deviation is that list_memories includes a direct object while remember, recall, forget, and reflect are verb-only.
Five tools is well-scoped for a memory server, and each tool earns its place in the set. It sits comfortably within the ideal 3–15 range with no redundant utilities.
The set covers the full memory lifecycle: create (remember), read (recall/list), delete (forget), and maintenance (reflect). No obvious dead ends or necessary operations are missing for the intended use case.