Skip to main content
Glama

save_memory

Save durable project knowledge to the active ContextDrop project. Use this proactively when meaningful project-specific knowledge emerges during a normal conversation, without waiting for the user to explicitly ask you to save it. Good candidates include established facts, definitions, architectural decisions, constraints, requirements, conventions, preferences, workflows, and other information that is likely to be useful in future AI sessions. Do not save casual conversation, generic knowledge, one-off questions, temporary debugging state, intermediate reasoning, speculation, short-lived tasks, duplicate information, or secrets/credentials. Before saving, extract the smallest useful durable statement and assign the appropriate category and memory_type. ContextDrop compares the candidate with existing memories and determines whether it should be created, updated, or ignored. Do not create duplicate memories.

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": {
      "title": "<title>",
      "content": "<content>",
      "category": "<category>",
      "memory_type": "<memory_type>"
    },
    "name": "save_memory"
  }
}

Input Schema

NameRequiredDescriptionDefault
titleYes
contentYes
categoryYes
memory_typeYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changed
    • addedInput schema / properties / category / enum
      Added value: +[
      +  "Technical",
      +  "Architecture",
      +  "Product",
      +  "Security",
      +  "Business",
      +  "Customer",
      +  "Content",
      +  "Operations",
      +  "Personal"
      +]
    • removedInput schema / properties / category / maxLength
      Removed value: -100
    • removedInput schema / properties / category / minLength
      Removed value: -1
    • removedInput schema / properties / confirm_existing
      Removed value: -{
      -  "default": false,
      -  "type": "boolean"
      -}
    • addedInput schema / properties / memory_type
      Added value: +{
      +  "enum": [
      +    "Fact",
      +    "Definition",
      +    "Decision",
      +    "Constraint",
      +    "Requirement",
      +    "Convention",
      +    "Preference",
      +    "Workflow",
      +    "Reference"
      +  ],
      +  "type": "string"
      +}
    • changedInput schema / required
      Previous value: -[
      -  "title",
      -  "content"
      -]New value: +[
      +  "title",
      +  "content",
      +  "category",
      +  "memory_type"
      +]
  2. First observed

TDQS

A4.1/5.0
Behavior4/5

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

With no annotations supplied, the description carries the full behavioral burden and does well: it discloses that the tool deduplicates against existing memories and that the outcome may be create, update, or ignore, which an agent could not infer from the schema. It leaves out response shape, permission/auth requirements, and any rate or size behavior beyond the schema's maxLength.

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

Conciseness4/5

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

The description is front-loaded with purpose, then qualification rules, then workflow; nearly every sentence contributes a decision rule. The only waste is mild redundancy between 'Do not save... duplicate information' and the closing 'Do not create duplicate memories.'

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 four-required-parameter write tool with no output schema and no annotations, it covers proactivity, exclusions, dedup handling, and the create/update/ignore outcome. What remains thin is guidance on category/memory_type selection and what the tool returns on success.

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 0%, so the description must compensate, and it only partly does: it names the required act of assigning 'category' and 'memory_type' and advises extracting 'the smallest useful durable statement' for content, but never explains how to choose among the nine enum values or how title should differ from content. The enum lists in the schema provide the candidate values, but selection criteria remain undocumented.

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 description opens with a specific verb and resource plus target scope ('Save durable project knowledge to the active ContextDrop project'), which is clearly distinct from search_memory, update_memory, get_project_context, and ping. It stops short of explicitly naming which sibling to use instead, but the intent is unambiguous.

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 gives explicit when-to-use guidance ('Use this proactively when meaningful project-specific knowledge emerges... without waiting for the user to ask') and an unusually detailed when-not-to-use list covering casual conversation, generic knowledge, debugging state, speculation, duplicates, and secrets. The contrast with update_memory is handled by noting the server decides create/update/ignore.

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.

Resources