Skip to main content
Glama

Client Decision Memory

get_client_memory
Read-onlyIdempotent

Recall recorded X1 decision memory for your own record or an assigned client, with category filtering, timestamps, status, owner labels, confidence, and provenance. When a saved decision includes recordUrl, use that exact link to open it in X1; never construct one from an id. Links to original documents are separate. When a member asks to recall a choice they saved or undid in their own household, first inspect one recent inventory with limit 10, without question or an unrequested category: saved wording may omit conversational names. Inventory entries are candidates, not named matches. Use at most one focused subject lookup if a related entry is missing or other related conditions are needed. Other decision questions can use a focused lookup directly. Reuse fresh same-scope results; an empty lexical search does not prove no decision exists. Never retry a denied read. Set drafts to authored_by_me only when explicitly reading the connected professional's own pending drafts; every other professional draft remains private. Scoped specialists are excluded.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "get_client_memory"
  }
}

Input Schema

NameRequiredDescriptionDefault
limitNoMaximum number of memory entries to return, capped at 50.
draftsNoSet to authored_by_me to include only pending decision drafts written by the connected user. Drafts from every other author remain private.
categoryNoOptional decision-memory category filter.
clientIdNoAssigned client or member user ID. Members may omit it for their own record; professionals must pass it unless X1 already supplied a client-scoped context.
questionNoOptional focused subject and explicitly requested period for a lexical history lookup; do not paste a compound conversational question. For a member's own saved or undone choice, inspect a recent inventory without question first unless fresh same-scope evidence already answers the recall.
checkLaterCopiesForDecisionIdNoFor one own saved decision returned by this read, optionally check for later copies of its original documents stored in X1. This does not update the decision or verify changed wording.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / checkLaterCopiesForDecisionId
      Added value: +{
      +  "description": "For one own saved decision returned by this read, optionally check for later copies of its original documents stored in X1. This does not update the decision or verify changed wording.",
      +  "maxLength": 128,
      +  "minLength": 1,
      +  "pattern": "^[A-Za-z0-9_-]+$",
      +  "type": "string"
      +}
  2. Changed1 schema field changed
    • changedInput schema / properties / question / description
      Previous value: -"Question used to retrieve relevant recorded work before the response limit."New value: +"Optional focused subject and explicitly requested period for a lexical history lookup; do not paste a compound conversational question. For a member's own saved or undone choice, inspect a recent inventory without question first unless fresh same-scope evidence already answers the recall."
  3. Changed1 schema field changed
    • addedInput schema / properties / question
      Added value: +{
      +  "description": "Question used to retrieve relevant recorded work before the response limit.",
      +  "maxLength": 1000,
      +  "minLength": 1,
      +  "type": "string"
      +}
  4. First observed

TDQS

A4.1/5.0
Behavior5/5

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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, yet the description adds genuinely new behavior: recordUrl is the only valid link and must not be reconstructed, original-document links are separate, inventory entries are candidates rather than matches, an empty lexical search does not prove absence, denied reads must not be retried, and other authors' drafts stay private. These are exactly the operational facts annotations cannot express.

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 purpose is front-loaded, but the body is a dense run of stacked imperatives with little paragraphing ('inspect one recent inventory with limit 10, without question or an unrequested category'), and several clauses restate the schema descriptions for question and drafts. Every idea is relevant, but the packing makes it harder to parse than it needs to be.

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, no-required-parameter read tool with annotations and no output schema, the definition covers privacy boundaries, link trust, caching/freshness, and retry behavior. What is thin is the return shape — it enumerates facets but does not describe entry structure, ordering, or what an empty result contains, which matters more here because no output schema exists.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3, but the description adds real intent beyond the schema: limit 10 on a first inventory sweep, no question or category on that first pass, drafts='authored_by_me' only for explicitly reading the connected professional's own pending drafts, and question should not be a pasted compound conversational question. That meaningfully upgrades how the parameters should be populated.

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 sentence gives a specific verb and resource ('Recall recorded X1 decision memory') plus the scope ('your own record or an assigned client') and the returned facets (category, timestamps, status, owner labels, confidence, provenance). It is clearly distinct from the neighbouring get_client_activity/profile/liquidity tools by resource, though it never explicitly contrasts itself with a named alternative.

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

Usage Guidelines4/5

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

Guidance is unusually detailed on mode selection: an inventory read with limit 10 for a member recalling their own saved/undone choice, at most one focused lookup otherwise, and focused lookup directly for other decision questions. It also states prohibitions (never construct a recordUrl, never retry a denied read, drafts only for the professional's own pending drafts). What it lacks is routing to any sibling tool, so the agent is guided within this tool but not across the toolset.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.