mnemostack

mnemostack_remember

Store a tenant-scoped memory with server-side embedding. Deterministic source/offset/text IDs make retries return duplicate, and chunked mode saves long documents as recallable chunks.

Instructions

Store a memory for the caller's tenant (the write counterpart of mnemostack_search).

A write tool: requires the write scope, embeds server-side, and stamps the process key's tenant so the memory lands in — and is recallable from — exactly this tenant's scope. Ids are deterministic from (source, offset, text): retries and repeated content return duplicate without a second embedding call. Long documents pass chunk=true and yield one result per chunk. Use mnemostack_invalidate to retract a stored memory.

Input Schema

TableJSON Schema

Name	Required	Description
`tags`	No	Optional tags stored in the payload.
`text`	Yes	The memory content to store. Embedded server-side. Plain items are capped at 32768 characters; longer documents need chunk=true.
`chunk`	No	Split a long document server-side into the same fixed character windows `mnemostack index` uses (identical chunk ids). Requires a non-empty source.
`offset`	No	Position within `source` for multi-part documents.
`source`	No	Logical origin (e.g. 'chat/2026-08-19'). With `offset` it forms the deterministic id: re-sending the same content is a no-cost duplicate, never a second copy.
`metadata`	No	Free payload fields, filterable at recall. Server-reserved keys (underscore-prefixed, structural ones like tenant_id/source, tags/timestamp — use their dedicated parameters — and the lifecycle marker invalidated_at, settable only via mnemostack_invalidate) are rejected.
`timestamp`	No	Event time of the content (ISO-8601); drives temporal recall.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Addedv2.3.1Aug 24, 2026

TDQS

A5/5.0

Behavior5/5

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

No annotations are present, so the description carries the full behavioral burden. It discloses the required write scope, server-side embedding, tenant stamping, deterministic duplicate behavior without a second embedding call, and chunked output behavior.

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

Conciseness5/5

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

Every sentence delivers distinct high-value information: scope, tenancy, idempotency, chunking, and the invalidation alternative. The primary action is front-loaded and there is no filler.

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

Completeness5/5

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

For a state-changing tool with seven parameters and no annotations, the description supplies crucial context: tenant isolation, permissions, duplicate semantics, chunking behavior, and the companion invalidation tool. Along with the schema, it is sufficient for correct invocation.

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

Parameters5/5

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

The schema already documents each parameter, but the description adds deeper semantics: source, offset, and text combine into a deterministic id, and chunk=true maps to one result per fixed character window. This is beyond the schema's parameter descriptions.

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

Purpose5/5

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

The description leads with a specific verb and resource: 'Store a memory for the caller's tenant,' and immediately labels it as the write counterpart of mnemostack_search. That alone lets an agent distinguish it from the search and invalidation siblings.

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

Usage Guidelines5/5

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

It explicitly tells the agent to use mnemostack_search for recall and mnemostack_invalidate for retraction, positioning this tool as the write path. It also provides a conditional usage rule for long documents via chunk=true.

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

Deploy Server