flashmind
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
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| memory_contextA | (Engram fallback.) Call ONCE at the start of a task when the mem_* tools are not available. Returns a digest of key memories (user profile, preferences, active projects, decisions) so you do not re-ask what is already known. |
| memory_searchA | Search memories by keyword. Call this BEFORE answering anything about the user, their projects, or things that may have been discussed before. If nothing is found, say you do not know — do not make it up. |
| memory_saveA | Save a durable fact: user identity/preferences, project details, decisions, lessons from mistakes. Never store your guesses as 'verified'. Never store passwords, API keys, tokens, card/account numbers, or ID numbers. |
| memory_getB | Read one full memory by id. |
| memory_listC | List memories (newest first), optionally filtered by category. |
| memory_updateC | Update an existing memory (e.g. a fact changed). Prefer updating over creating duplicates. |
| memory_deleteB | Delete a memory. Only when the user asks or the fact is proven wrong. |
| memory_verifyA | Check a claim against memory before stating it as fact. Returns candidate memories and a status: SUPPORTED_CANDIDATES (compare the content yourself) or NO_EVIDENCE. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| anti_hallucination_rules | Working rules: use memory, do not make things up, cite sources and confidence. |
| recall | Recall everything known about a topic. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
The tools are mostly distinct: save, get, list, update, delete form clear CRUD roles, while context/search/verify have explicit usage guidance. There is some overlap between search, list, context, and verify since all retrieve candidate memories, but the descriptions give enough cues to choose correctly.
All tools use a consistent memory_ prefix followed by a clear verb or noun, in uniform snake_case. The pattern is predictable across the entire set.
Eight tools is well-scoped for a memory service: CRUD operations plus search, list, context digest, and verification. No tool feels redundant or missing at the count level.
The surface covers the full memory lifecycle: create/save, read (get/list/search/context), update, delete, and verification. No obvious operational gaps for the stated purpose.