Skip to main content
Glama

Remember

padam_write

Store persistent memories across sessions while resolving duplicates or contradictions, preserving history, masking secret-looking values, and holding sensitive writes for owner approval.

Instructions

Store something worth remembering across sessions: a stated preference, a decision, a stable fact, a correction. If a close record already exists, PADAM resolves the relation itself (duplicate, refinement or contradiction) and keeps the previous version in history. Secret-looking values are masked before storing. A write that looks like an instruction, or that would replace a protected identity or preference record, is held for the owner's review and does not take effect until the owner approves it.

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": {
      "content": "<content>"
    },
    "name": "padam_write"
  }
}

Input Schema

NameRequiredDescriptionDefault
kindNoType governs how fast it fades: preference and identity never fade, decision and correction over a year, fact over half a year, state over two weeks, event over two days. Omit to detect automatically.
linksNoOptional typed graph links to existing records (cause, correction, contradiction, source)
scopeNoMemory scope, e.g. 'project:code'global
originNoWhere this came from. user: the person said it. agent: your own conclusion. external: read from a web page, file or tool output.agent
contentYesThe record, as one clear sentence
agent_idNoIdentifier of the agent storing the record (e.g. claude-code, cursor)
valid_toNoOptional end of validity window (ISO date or UTC timestamp)
dedup_modeNoOptional deduplication strategy: atomic_merge detects close active facts in the same scope and links refinements cleanly while preserving quarantine guards.default
importanceNo
valid_fromNoOptional start of validity window (ISO date or UTC timestamp)
visibilityNoOptional workspace visibility: private, team_read, or team_write
derived_fromNoIds of the records this conclusion was drawn from. When the person forgets a source, its conclusions can be forgotten with it.
workspace_idNoOptional workspace ID when storing team-shared memories

Output Schema

NameRequiredDescriptionDefault
idYes
outcomeYes
dlpAliasesNo
holdReasonNo
mergedWithIdNo
maskedSecretsYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed8 schema fields changedv0.6.2
    • addedInput schema / properties / agent_id
      Added value: +{
      +  "description": "Identifier of the agent storing the record (e.g. claude-code, cursor)",
      +  "maxLength": 64,
      +  "minLength": 1,
      +  "type": "string"
      +}
    • addedInput schema / properties / dedup_mode
      Added value: +{
      +  "default": "default",
      +  "description": "Optional deduplication strategy: atomic_merge detects close active facts in the same scope and links refinements cleanly while preserving quarantine guards.",
      +  "enum": [
      +    "default",
      +    "atomic_merge"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / links
      Added value: +{
      +  "description": "Optional typed graph links to existing records (cause, correction, contradiction, source)",
      +  "items": {
      +    "additionalProperties": false,
      +    "properties": {
      +      "direction": {
      +        "enum": [
      +          "in",
      +          "out"
      +        ],
      +        "type": "string"
      +      },
      +      "id": {
      +        "maxLength": 64,
      +        "minLength": 1,
      +        "type": "string"
      +      },
      +      "relation": {
      +        "enum": [
      +          "cause",
      +          "correction",
      +          "contradiction",
      +          "source",
      +          "derived"
      +        ],
      +        "type": "string"
      +      },
      +      "source": {
      +        "maxLength": 64,
      +        "minLength": 1,
      +        "type": "string"
      +      },
      +      "target": {
      +        "maxLength": 64,
      +        "minLength": 1,
      +        "type": "string"
      +      },
      +      "type": {
      +        "enum": [
      +          "cause",
      +          "correction",
      +          "contradiction",
      +          "source",
      +          "derived"
      +        ],
      +        "type": "string"
      +      }
      +    },
      +    "type": "object"
      +  },
      +  "maxItems": 20,
      +  "type": "array"
      +}
    • addedInput schema / properties / valid_from
      Added value: +{
      +  "description": "Optional start of validity window (ISO date or UTC timestamp)",
      +  "maxLength": 40,
      +  "pattern": "^\\d{4}-\\d{2}-\\d{2}(T\\d{2}:\\d{2}(:\\d{2}(\\.\\d{1,6})?)?(Z|[+-]\\d{2}:\\d{2})?)?$",
      +  "type": "string"
      +}
    • addedInput schema / properties / valid_to
      Added value: +{
      +  "description": "Optional end of validity window (ISO date or UTC timestamp)",
      +  "maxLength": 40,
      +  "pattern": "^\\d{4}-\\d{2}-\\d{2}(T\\d{2}:\\d{2}(:\\d{2}(\\.\\d{1,6})?)?(Z|[+-]\\d{2}:\\d{2})?)?$",
      +  "type": "string"
      +}
    • addedInput schema / properties / visibility
      Added value: +{
      +  "description": "Optional workspace visibility: private, team_read, or team_write",
      +  "enum": [
      +    "private",
      +    "team_read",
      +    "team_write"
      +  ],
      +  "type": "string"
      +}
    • addedInput schema / properties / workspace_id
      Added value: +{
      +  "description": "Optional workspace ID when storing team-shared memories",
      +  "maxLength": 64,
      +  "minLength": 1,
      +  "type": "string"
      +}
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
      +  "properties": {
      +    "dlpAliases": {
      +      "items": {
      +        "type": "string"
      +      },
      +      "type": "array"
      +    },
      +    "holdReason": {
      +      "type": [
      +        "string",
      +        "null"
      +      ]
      +    },
      +    "id": {
      +      "type": "string"
      +    },
      +    "maskedSecrets": {
      +      "type": "integer"
      +    },
      +    "mergedWithId": {
      +      "type": [
      +        "string",
      +        "null"
      +      ]
      +    },
      +    "outcome": {
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "id",
      +    "outcome",
      +    "maskedSecrets"
      +  ],
      +  "type": "object"
      +}
  2. First observedv0.4.8

TDQS

A4.1/5.0
Behavior5/5

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

With annotations covering only the safety profile, the description contributes real behavior: near-duplicate records are resolved automatically with the prior version kept in history, secret-looking values are masked before storage, and instruction-like or protected-identity writes are quarantined for owner approval. These are non-obvious operational traits an agent cannot infer from readOnlyHint/destructiveHint=false.

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?

Four sentences, front-loaded with the core action and then progressively disclosing dedup, masking, and quarantine behavior. Each sentence earns its place; no filler.

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 13-parameter mutation tool with rich schema coverage, an output schema, and annotations, the description covers the behaviors that matter (history retention, masking, review holds). The remaining parameters are handled by the schema, so nothing critical is missing.

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 coverage is 92%, so the schema already documents all 13 parameters including kind enums and dedup_mode. The description adds only marginal mapping by giving example record types that correspond to the kind enum and alluding to dedup resolution; it adds nothing about scope, origin, validity windows, or visibility.

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?

States a specific verb-plus-resource ('Store something worth remembering across sessions') and enumerates the record types it accepts, so an agent can tell this is the memory-write tool. It stops short of naming sibling operations such as padam_confirm or padam_refute to sharpen the boundary.

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?

Gives clear context on what qualifies for storage (preference, decision, fact, correction) and explains the dedup path when a close record exists. It does not explicitly say when to prefer padam_confirm or padam_refute instead, so the routing guidance is contextual rather than exhaustive.

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