Nexus Memory
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| JINA_API_KEY | No | Jina API key for embeddings (1024d) | |
| GOOGLE_API_KEY | No | Google API key for Vertex AI embeddings (768d) | |
| OPENAI_API_KEY | No | OpenAI API key for embeddings (1536d) | |
| VOYAGE_API_KEY | No | Voyage API key for embeddings (1024d) |
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
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| rememberB | Store a memory for AI agents. Persists information across sessions. |
| recallA | Search memories. Returns relevant context from past sessions. |
| forgetA | Delete a specific memory by ID. |
| updateB | Update an existing memory in-place without losing metadata. |
| fact_historyA | Trace the supersession chain of a memory: the point itself plus all successors (and predecessors via superseded_by), ordered by valid_from. Shows how a fact evolved over time. |
| healthA | Check if Nexus Memory is running and healthy. |
| check_updateB | Check if a newer version is available on GitHub. |
| do_updateC | Update Nexus Memory to the latest version. Pulls from GitHub and reinstalls. |
| subscribeA | Register a webhook URL to receive HTTP POST notifications when a memory event of the given type fires. Returns the subscription id (UUID) which you need to unsubscribe. Subscriptions are stored in ~/.nexus-webhooks.json and survive server restarts. |
| unsubscribeA | Remove a webhook subscription by its id (returned from subscribe). |
| list_subscriptionsA | List all currently registered webhook subscriptions (id, event_type, webhook_url, created_at). |
| backupA | Create a full backup of all memories as JSON file. Includes payloads + vectors. Saved to ~/.nexus-memory/backups/. Runs automatically every 24h - use this for manual backup on demand. |
| restoreA | Restore memories from a backup JSON file. By default reuses stored vectors (zero API cost). Set reembed=true to re-embed with current provider (for provider changes). |
| guardrail_checkA | Active Guardrails: Check if an action is safe before executing it. Queries Nexus Memory for protection rules (category='rule' with protection keywords like 'niemals', 'never delete', 'protected'). Returns allow/block/override verdict. Use before destructive operations (rm, drop, kill, overwrite, recreate). |
| guardrail_overrideA | Active Guardrails: Record a guardrail override with full audit trail. Required when guardrail_check returns 'block' but the action is explicitly authorized. The override is stored as a private session memory entry for audit. Requires explicit reasoning (min 10 chars). |
| graph_traverseA | Knowledge Graph: Multi-hop traversal from a starting fact. Answers 'what is connected to X?' across the entity graph. Returns a list of {fact_id, depth, relation, path} dicts. |
| find_entitiesA | Knowledge Graph: Find all entity-typed memories in Qdrant. Returns a list of {id, name, entity_type, content, attributes} dicts. |
| get_subgraphA | Knowledge Graph: Get a subgraph centered on a fact for visualization. Returns {nodes, edges} where nodes have {id, depth} and edges have {source, target, relation}. |
| get_relatedA | Knowledge Graph: Get directly related facts (1-hop, bidirectional). Returns a list of {fact_id, relation, edge_id, direction} dicts. |
| cost_routing_statsA | Cost-Aware Routing: Get statistics about embedding provider routing. Shows available providers, routing decisions per provider, and tier configuration. |
| cost_routing_explainA | Cost-Aware Routing: Explain the routing decision for a memory category. Returns a human-readable explanation of which provider will be used and why. |
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 21 tools
Most tools are clearly distinct (remember/recall/forget/update, subscribe/unsubscribe/list_subscriptions, backup/restore, guardrail_check/guardrail_override, graph_traverse/find_entities/get_subgraph/get_related, cost_routing_stats/cost_routing_explain). The only potential confusion is check_update vs do_update (checking for an update vs performing the update), but their descriptions make the distinction clear.
The naming is mostly verb_noun (check_update, remember, recall, forget, update, subscribe, unsubscribe, list_subscriptions, backup, restore, find_entities, get_subgraph, get_related), but there are inconsistencies: 'do_update' is an odd variant of 'update', and 'fact_history' and 'health' are noun-only rather than verb_noun. The mixed conventions are still readable and mostly predictable.
21 tools is on the higher end but appropriate for the server's broad scope: memory CRUD, webhooks, backup/restore, guardrails, knowledge graph, cost routing, and system maintenance. Each tool serves a distinct function, though the count is slightly heavy and could feel overwhelming.
The memory lifecycle is well covered (remember, recall, update, forget, fact_history), and the server includes thoughtful extras like backup/restore, guardrails, and webhooks. Minor gaps exist: there's no explicit 'list all memories' tool, and the guardrail system has check/override but no way to create or manage protection rules directly.