Skip to main content
Glama

Remember a durable fact

brain_remember

Append a durable owner fact—preference, correction, project constraint, or reference—to the memory file so it informs brain_identity and can replace outdated facts.

Instructions

Append one durable fact about the owner to their memory file: a preference, a correction on how to work, a project constraint, or a reference. It becomes part of brain_identity immediately. Include why it matters and how to apply it. When it changes an earlier fact, pass replaces with that fact's id (the stamp in its heading, e.g. '2026-09-22 18:21'): the old fact stays in the file as history and leaves brain_identity. For a fact with one current value, pass subject and attribute (e.g. 'JobMail' and 'database'); a later fact with the same pair replaces it on its own. Use valid_until for something temporary and valid_from when it became true earlier than now. Near-duplicates of current facts are refused, and a fact that looks like it changes a current one is held back until you say whether it replaces it; pass force=true to store it as is. Do not store session trivia.

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": {
      "fact": "<fact>",
      "category": "<category>"
    },
    "name": "brain_remember"
  }
}

Input Schema

NameRequiredDescriptionDefault
whyNo
factYes
forceNoStore even if it looks like a duplicate or a change to a current fact
subjectNoWhat the fact is about, e.g. 'JobMail' or 'the owner'. Pass with attribute.
categoryYesuser = who they are; feedback = how to work with them; project = ongoing work; reference = pointer
replacesNoIds of current facts this one makes outdated, e.g. ['2026-09-22 18:21']
attributeNoWhich property of the subject, e.g. 'database' or 'employer'. Pass with subject.
valid_fromNoWhen it became true, if earlier or later than now: YYYY-MM-DD or YYYY-MM-DD HH:MM (UTC)
valid_untilNoWhen it stops being true, for temporary facts: YYYY-MM-DD (holds through that day) or YYYY-MM-DD HH:MM (UTC)
how_to_applyNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed6 schema fields changedv0.5.2
    • addedInput schema / properties / attribute
      Added value: +{
      +  "description": "Which property of the subject, e.g. 'database' or 'employer'. Pass with subject.",
      +  "maxLength": 120,
      +  "minLength": 1,
      +  "type": "string"
      +}
    • changedInput schema / properties / force / description
      Previous value: -"Store even if it looks like a duplicate"New value: +"Store even if it looks like a duplicate or a change to a current fact"
    • addedInput schema / properties / replaces
      Added value: +{
      +  "description": "Ids of current facts this one makes outdated, e.g. ['2026-09-22 18:21']",
      +  "items": {
      +    "maxLength": 30,
      +    "type": "string"
      +  },
      +  "maxItems": 10,
      +  "type": "array"
      +}
    • addedInput schema / properties / subject
      Added value: +{
      +  "description": "What the fact is about, e.g. 'JobMail' or 'the owner'. Pass with attribute.",
      +  "maxLength": 120,
      +  "minLength": 1,
      +  "type": "string"
      +}
    • addedInput schema / properties / valid_from
      Added value: +{
      +  "description": "When it became true, if earlier or later than now: YYYY-MM-DD or YYYY-MM-DD HH:MM (UTC)",
      +  "maxLength": 40,
      +  "type": "string"
      +}
    • addedInput schema / properties / valid_until
      Added value: +{
      +  "description": "When it stops being true, for temporary facts: YYYY-MM-DD (holds through that day) or YYYY-MM-DD HH:MM (UTC)",
      +  "maxLength": 40,
      +  "type": "string"
      +}
  2. First observedv0.4.0

TDQS

A4.2/5.0
Behavior3/5

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

Strong disclosure of non-obvious behavior that annotations cannot express: near-duplicates are refused, apparent changes are held back pending confirmation, superseded facts remain in the file as history but leave brain_identity, and new facts take effect immediately. However it does not describe what a call returns (e.g. the new fact's id), which an agent needs for later `replaces` calls; the hint that ids appear as heading stamps is indirect.

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?

Purpose is front-loaded in the first clause, and nearly every sentence carries a rule or condition. It is dense to the point of being a wall of rules, with parameter guidance interleaved rather than grouped, which slightly hurts scanning.

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?

Covers the full write-path behavior for a 10-parameter, non-idempotent mutation with no output schema: dedupe, history retention, temporal bounds, and override. The one gap is the response shape and how to obtain a fact id, which matters because `replaces` depends on it.

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?

Adds genuine meaning beyond the 70%-covered schema: the `replaces` value format and its history-preservation effect, the subject+attribute pairing that makes later facts self-replacing, and the temporal semantics of valid_from/valid_until. Even the undocumented `why` and `how_to_apply` fields are given intent.

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?

Specific verb and resource ('Append one durable fact about the owner to their memory file') plus an enumeration of the kinds of facts it accepts, which pins down scope precisely. It does not name a sibling to route against (e.g. brain_capture vs brain_write), so differentiation is implied rather than stated.

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?

Explicit conditions for each mode of use: pass `replaces` when superseding an earlier fact, use subject/attribute for single-valued facts, `valid_until` for temporary facts, `valid_from` for backdating. It also gives an explicit exclusion ('Do not store session trivia') and explains the force override path.

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