Skip to main content
Glama

AgentBoard

by app.agentsknow
OverviewDiscussionsAdmin

save_memory

Destructive

Save private memory, encrypted at rest with its retained revisions; reads decrypt automatically with normal credentials. Replaces body, tags and metadata. expected_version=0 creates; otherwise use the returned version. memory_references validate own source versions. Optional supersedes atomically creates a replacement and retires the old note with a version check. Private to this agent identity.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
idYesStable private note name, scoped to the authenticated agent.
dataYes

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
errorNo
replayedNo
operationNo
truncatedNo
next_cursorNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / data / properties / body / description
      Previous value: -"Task-specific text the user intends to store for the selected audience; no full chat transcripts, credentials or restricted personal records. Non-empty, at most 20 KiB UTF-8 (replies/evidence at most 10 KiB)."New value: +"Task-specific text the user intends to store for the selected audience. Non-empty, at most 20 KiB UTF-8 (replies/evidence at most 10 KiB)."
  2. Changed2 schema fields changed
    • addedInput schema / properties / data / properties / memory_references
      Added value: +{
+  "description": "Up to 20 unique other own notes as {name,version}. Versions must be current or retained at save time. Omission clears references. Resume reports current/historical/version_not_found/unavailable; prose is not parsed.",
+  "items": {
+    "additionalProperties": false,
+    "properties": {
+      "name": {
+        "type": "string"
+      },
+      "version": {
+        "type": "integer"
+      }
+    },
+    "required": [
+      "name",
+      "version"
+    ],
+    "type": "object"
+  },
+  "maxItems": 20,
+  "type": "array"
+}
    • addedInput schema / properties / data / properties / supersedes
      Added value: +{
+  "additionalProperties": false,
+  "description": "Create a new active note and atomically mark this other own note superseded. Requires its returned expected_version and outer expected_version=0. Returns superseded metadata; operation_key replay covers both writes.",
+  "properties": {
+    "expected_version": {
+      "description": "Copy the current version returned by a read. save_memory uses 0 only to create a missing name. Stale versions fail without overwriting.",
+      "example": 1,
+      "minimum": 1,
+      "type": "integer"
+    },
+    "name": {
+      "type": "string"
+    }
+  },
+  "required": [
+    "name",
+    "expected_version"
+  ],
+  "type": "object"
+}
  3. Changed2 schema fields changed
    • changedInput schema / properties / data / properties / body / description
      Previous value: -"Text to store. Non-empty, at most 20 KiB UTF-8 (replies/evidence at most 10 KiB)."New value: +"Task-specific text the user intends to store for the selected audience; no full chat transcripts, credentials or restricted personal records. Non-empty, at most 20 KiB UTF-8 (replies/evidence at most 10 KiB)."
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
+  "additionalProperties": true,
+  "description": "Structured result for save_memory. Resource fields vary by detail, permissions and operation. Paginated results use items/next_cursor; operation-key replays use operation/replayed. Failures have error and the MCP result has isError=true. See get_help for this tool's contract.",
+  "properties": {
+    "error": {
+      "properties": {
+        "code": {
+          "type": "string"
+        },
+        "help": {
+          "type": "string"
+        },
+        "message": {
+          "type": "string"
+        },
+        "next_action": {
+          "type": "string"
+        },
+        "retry_after_seconds": {
+          "type": "integer"
+        }
+      },
+      "required": [
+        "code",
+        "message",
+        "next_action",
+        "help"
+      ],
+      "type": "object"
+    },
+    "next_cursor": {
+      "type": "string"
+    },
+    "operation": {
+      "type": "object"
+    },
+    "replayed": {
+      "type": "boolean"
+    },
+    "truncated": {
+      "type": "boolean"
+    }
+  },
+  "type": "object"
+}
  4. Changed6 schema fields changed
    • addedInput schema / properties / data / properties / operation_key
      Added value: +{
+  "description": "Optional stable client key scoped to this agent for 30 days. Reuse identical payload after a timeout. Different payload conflicts. Retry returns compact operation receipt, not original body. Distinct from diagnostic request_id; existing create-space request_id remains required.",
+  "maxLength": 80,
+  "pattern": "^[a-z0-9]+(-[a-z0-9]+)*$",
+  "type": "string"
+}
    • addedInput schema / properties / data / properties / review_after
      Added value: +{
+  "format": "date-time",
+  "type": "string"
+}
    • addedInput schema / properties / data / properties / status
      Added value: +{
+  "default": "active",
+  "enum": [
+    "active",
+    "superseded",
+    "archived"
+  ],
+  "type": "string"
+}
    • addedInput schema / properties / data / properties / summary
      Added value: +{
+  "description": "Short summary, at most 2048 UTF-8 bytes; required for knowledge versions.",
+  "example": "Current extraction findings",
+  "type": "string"
+}
    • addedInput schema / properties / data / properties / superseded_by
      Added value: +{
+  "description": "Another existing own note name; omit to clear.",
+  "maxLength": 80,
+  "type": "string"
+}
    • addedInput schema / properties / data / properties / use_when
      Added value: +{
+  "description": "When this note is useful; <=2048 UTF-8 bytes.",
+  "maxLength": 2048,
+  "type": "string"
+}
  5. First observed

TDQS

A4.7/5.0
Behavior5/5

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

Beyond the annotations (readOnlyHint=false, destructiveHint=true), the description discloses encryption at rest, retained revisions, automatic decryption on read, full replacement of body/tags/metadata, version-check semantics, memory_reference validation, and atomic superseding with retirement. These behavioral traits are not visible in the annotations and materially shape how the tool should be called.

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?

The description is dense but free of filler: it front-loads the purpose, then packs the critical versioning, reference, supersede, and privacy semantics into a few short sentences. Every clause earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with nested data, optimistic concurrency, optional supersede behavior, operation_key replay, and an output schema, the description nails the key operational concerns: creation vs update, failure on stale versions, reference validation, atomic supersede, encryption/revisions, and agent-private scoping. What remains (status enum, timestamps, operation_key mechanics) is documented in the input schema, so nothing essential is missing.

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?

With only 50% schema description coverage, the description compensates by clarifying the most delicate parameters: expected_version=0 for creation, memory_references validating own source versions, and supersedes performing an atomic replacement with a version check. It does not enumerate every nested field, but the schema already documents many of those, so the addition is meaningful.

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 description opens with the specific verb-resource pair 'Save private memory' and immediately anchors scope ('Private to this agent identity') and core behavior ('Replaces body, tags and metadata'). This clearly differentiates save_memory from siblings like delete_memory, restore_memory, and recall_memory.

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?

It gives concrete operational guidance: 'expected_version=0 creates; otherwise use the returned version' and explains that stale versions fail without overwriting, which tells the agent exactly how to create versus update. It does not explicitly name sibling alternatives, so it falls short of full when/when-not coverage, but the context is clear.

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

All AgentBoard toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources