Skip to main content
Glama

remember

Store an append-only memory with tags, importance, source, and provenance so it can be recalled, updated via supersession key, or erased with its lineage.

Instructions

Store a memory (append-only; raw text is never edited afterward). tags group memories into cohorts; value (>=1) is its importance — higher-value memories outrank merely-similar ones at recall, and recall itself nudges value up. mtype ∈ {episodic, semantic, procedural} sets the decay prior — episodic (events) fades fast, semantic (durable facts) slow, procedural (rules / preferences) barely; pass it when you know the kind, else it's inferred.

Optional key is a deterministic (subject, relation) supersession key (e.g. "billing-api::auth-method"): storing a new value with the same key retires the old one so recall never returns the stale value — no similarity threshold, no extra LLM call. Use it for facts that get updated (config, prices, versions, status). Pass object = the asserted VALUE (e.g. "frankfurt") alongside key: with the echo guard on (default here), a later RE-STATEMENT of an already-retired value cannot resurrect it (a corrected fact stays corrected even if the old value is said again). Without object the guard still catches a verbatim restatement (text hash), but a reworded one needs the value in object to be caught. Set reaffirm=True to intentionally revert to a previously-retired value (an explicit change-of-mind, not an echo).

source — WHO OR WHAT this came from ("crm/alice", "user-42", "docs.example.com/pricing"). Pass it whenever the memory is about, or came from, an identifiable person or system. It is what makes the memory reachable later by SUBJECT rather than only by id: forget_subject("crm/alice") erases a person's data and everything derived from it, erasure_audit can then say whether anything survived, and slash can forfeit a source's standing after a bad outcome. Without it a record is attributable to nothing, and none of those can reach it -- measured: a store written through this server answered would_erase=0 to every right-to-erasure request, while the same write with a source answered 1.

derived_from — the ids this memory was BUILT FROM (a summary, a merge, a conclusion drawn from earlier notes). Provenance rides along the edge: erasing the source erases what was derived from it, so a summary of a person's file goes when their file goes. erasure_audit walks these edges and reports unaudited -- never a pass -- when nothing declares them, because a store with no edges to walk has not been checked, it has been left uninspected.

If this server was started with a PROJECT scope (--project <name> / INSPEXIMUS_PROJECT), the memory is stamped with it and later recalls in OTHER projects will not return it. The active scope is echoed back as project in the result (null = unscoped, shared by every project).

Returns the new id.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
keyNo
tagsNo
textYes
mtypeNo
valueNo
objectNo
sourceNo
user_idNo
agent_idNo
reaffirmNo
session_idNo
derived_fromNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed5 schema fields changedv2.20.1
    • addedInput schema / properties / agent_id
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Agent Id"
      +}
    • addedInput schema / properties / derived_from
      Added value: +{
      +  "anyOf": [
      +    {
      +      "items": {
      +        "type": "string"
      +      },
      +      "type": "array"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Derived From"
      +}
    • addedInput schema / properties / session_id
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "Session Id"
      +}
    • addedInput schema / properties / source
      Added value: +{
      +  "default": "",
      +  "title": "Source",
      +  "type": "string"
      +}
    • addedInput schema / properties / user_id
      Added value: +{
      +  "anyOf": [
      +    {
      +      "type": "string"
      +    },
      +    {
      +      "type": "null"
      +    }
      +  ],
      +  "default": null,
      +  "title": "User Id"
      +}
  2. First observedv0.1.0

TDQS

A4.6/5.0
Behavior5/5

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

With no annotations provided, the description carries the full burden and delivers extensively: append-only immutability, the side effect that 'recall itself nudges value up', the echo guard, deterministic supersession with 'no similarity threshold, no extra LLM call', project scoping, and provenance cascade behavior. It even includes a measured evidence point contrasting would_erase=0 versus 1.

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?

Long but never padded: each paragraph covers one behavioral theme (metadata, supersession, attribution, provenance, scoping) with concrete examples like 'billing-api::auth-method' and 'frankfurt'. It is front-loaded with the core purpose and append-only invariant, and every sentence conveys a semantic that cannot be inferred from the parameter names or schema.

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 12-parameter tool with zero schema descriptions, no annotations, and no output schema, this description covers nearly everything an agent needs: return value ('Returns the new id'), project scoping, side effects, and parameter behavior. The remaining gaps are minor but real: user_id/agent_id/session_id are never explained, and error behavior (e.g., invalid mtype or value < 1) is unspecified.

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?

For nine of twelve parameters (tags, value, mtype, key, object, reaffirm, source, derived_from, text) the description provides semantics far beyond any schema field: value ranges (>=1), enumerated mtype values, and behavioral consequences like higher-value memories outranking merely-similar ones at recall. However, with 0% schema description coverage, the complete silence on user_id, agent_id, and session_id leaves three parameters semantically empty in both description and schema.

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 opening sentence 'Store a memory (append-only; raw text is never edited afterward)' states a specific verb and resource, and the append-only qualifier immediately frames the tool's core contract. The detailed discussion of recall outranking, supersession, and key-based retirement makes its role among 71 memory siblings unmistakable, even without explicitly naming alternatives.

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?

The description gives explicit when-to-use instructions at the parameter level: 'Use it for facts that get updated (config, prices, versions, status)' for key, 'pass it when you know the kind, else it's inferred' for mtype, and 'Pass it whenever the memory is about, or came from, an identifiable person or system' for source. What's missing is explicit routing against sibling tools like observe or remember_decision — there is no 'when not to use remember' statement.

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