rmbr
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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 |
|---|---|
| searchA | Search documents indexed via Index.add_files()/add_text() - the
agent's reference knowledge base, not its own memory. Use Read-only, no side effects. Results are ranked best-match-first by
a hybrid of keyword and semantic similarity, not necessarily
indexing order. If |
| recallA | Recall notes previously saved via Read-only, no side effects. Results are ranked best-match-first by
a hybrid of keyword and semantic similarity, not necessarily the
order they were remembered. If |
| rememberA | Save a new note to this agent's memory. Returns the new memory's id. Always inserts - calling this twice with the same text creates two separate memories, since this server doesn't have deduplication enabled by default. If the server was configured with a memory cap (max_memories), the oldest unpinned memory may be evicted to make room for this one. pinned=True exempts a memory from that eviction permanently - there's no way to un-pin it later through this tool. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| examples_index | List the available rmbr usage-pattern examples (see rmbr://examples/{pattern}). |
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: search retrieves from the knowledge base, recall retrieves from the agent's own memory, and remember writes a new memory. The descriptions explicitly cross-reference each other to prevent confusion.
All tool names are single, lowercase verbs (search, recall, remember) that directly describe their action. This is a consistent and predictable naming pattern.
With only three tools, the server is tightly scoped to its core purpose of storing and retrieving memories and reference documents. The count is appropriate for this narrow domain.
The server provides create and read operations for memories but lacks update and delete capabilities, leaving no way to correct or remove stored memories. Additionally, the knowledge base search has no matching tool to add or manage indexed documents, making the surface incomplete for full lifecycle management.