MCP tools reference
llm-mem-mcp exposes 32 tools to MCP clients. They’re grouped by purpose here.
The 32 tools at a glance
Storing
add_content_memory— raw verbatim textadd_intuitive_memory— AI-extracted structured factsstore_memories— batch storeupload_document— small documents in one shotingest— alternate ingest entry point
Finding
query_memory— full pyramid + graph searchsearch_memory— simple semantic searchlist_memories— filter-based browseget_memory— by ID
Chronology
get_timeline— bucketed listget_timeline_graph— nodes + edgesget_context_resume— progressive precision
Navigation & relations
navigate_memory— zoom in/out across pyramidupdate_memory— edit content or relationsforce_link— create a relationremove_relation— delete a relation
Abstraction
start_abstraction_pipeline— start background workersstop_abstraction_pipeline— pause workerstrigger_abstraction— force one specific abstractioncreate_abstraction— manual layer creation
Banks
list_memory_bankscreate_memory_bankrename_memory_bankbackup_bankrestore_bankcleanup_resources— delete models or banks
Documents
Introspection
system_status— health + token statshealth_check— config + live backend probecheck_consistency— DB integrityhelp— built-in tool guidance
How parameters work
Every tool takes a JSON object of parameters. Optional parameters can be omitted. Parameters marked as accepting a bank field default to the "default" bank if not provided.
The common bank parameter
Most tools accept an optional bank string. If omitted, the tool operates on the "default" bank.
{ "query": "JWT auth", "bank": "research" }
Agent and user IDs
| Field | Used by | Purpose |
|---|---|---|
agent_id | store/find tools | The agent storing or querying. Set once via --agent CLI flag and inherited by all calls. |
user_id | store tools | Optional. Only needed when multiple users share a bank. |
Common response shape
Tools return a JSON object with at least:
{
"success": true,
"message": "human-readable status",
"data": { /* tool-specific */ },
"error": null
}
On failure, success is false and error contains a human-readable description.
Reading the per-tool pages
Each category page has a one-paragraph description of the tool, the input schema (as a table), and the output schema. For the canonical schema, see src/operations/tools.rs in the source.