Skip to main content
Glama

memory_store

Store facts, notes, and preferences as persistent memories with optional tags and types. Save incremental entries by conversation ID to skip semantic deduplication.

Instructions

Store new information with optional tags.

                Accepts two tag formats in metadata:
                - Array: ["tag1", "tag2"]
                - String: "tag1,tag2"

                Use `conversation_id` to bypass semantic deduplication when saving
                incremental memories from the same conversation.

               Examples:
                # Using array format:
                {
                    "content": "Memory content",
                    "metadata": {
                        "tags": ["important", "reference"],
                        "type": "note"
                    }
                }

                # Using string format(preferred):
                {
                    "content": "Memory content",
                    "metadata": {
                        "tags": "important,reference",
                        "type": "note"
                    }
                }

                # Using conversation_id to save incremental notes:
                {
                    "content": "User prefers dark mode",
                    "conversation_id": "conv_abc123",
                    "metadata": {
                        "tags": "preference,ui"
                    }
                }

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "content": "<content>"
    },
    "name": "memory_store"
  }
}

Input Schema

NameRequiredDescriptionDefault
storeNoTarget store partition (default: 'default'). Use 'docs' for documents, 'all' for cross-store search.
contentYesThe memory content to store, such as a fact, note, or piece of information.
agent_idNoOptional authoring agent id (defaults to the MCP_AGENT_ID env var, else unset). Stored in metadata for multi-agent attribution; null means unknown.
metadataNoOptional metadata about the memory, including tags and type.
conversation_idNoOptional conversation identifier. When provided, semantic deduplication is skipped, allowing multiple incremental memories from the same conversation to be stored even if their content is topically similar. Exact duplicate hashes are still rejected.
restore_deletedNoBring back content that was deleted on this server. Default false: re-sending deleted content is refused with 'Not stored: this content was deleted on <date>; pass restore_deleted=true to bring it back'. Set true only when the user deliberately wants the deleted memory back.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv1.0.1

TDQS

A3.6/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only supply destructiveHint=false, so the description carries most of the behavioral load and does so well: it discloses semantic deduplication, that exact duplicate hashes are still rejected, that deleted content is refused with a specific message unless restore_deleted is set, and that unknown types are silently coerced to 'observation' with a warning. It stops short of describing the success response shape.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The behavioral notes are front-loaded and earn their place, but the three JSON examples are bloated: the array-format and string-format examples are near-identical, and the string one is even labeled 'preferred' without further explanation. The examples could be collapsed to one compact pair of snippets without losing meaning.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a six-parameter write tool with nested metadata and no output schema, the description covers the non-obvious failure modes (dedup, deleted-content refusal, type coercion) that an agent most needs to avoid errors. Remaining gaps are modest: no description of the successful return payload and no coverage of the store/agent_id parameters, though both are fully documented in the schema.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every parameter already carries its own documentation including the dual tag formats and conversation_id dedup behavior. The description's tag-format explanation and JSON examples largely restate what the schema already states, adding only concrete payload illustrations; baseline 3 applies when the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The opening line gives a specific verb and resource ('Store new information'), immediately telling an agent this is a write of a new memory rather than a search or update. It does not, however, differentiate itself from close siblings such as memory_store_session, memory_update, or memory_ingest, so the agent must infer the boundary.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage guidance exists but is parameter-scoped rather than tool-scoped: it explains when to pass conversation_id (incremental saves from the same conversation) and when to set restore_deleted=true (only when the user deliberately wants deleted content back). There is no statement of when to choose this tool over memory_store_session or memory_update, leaving tool selection implied.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.