Skip to main content
Glama

Dayze — Life Context

by com.dayze
OverviewDiscussionsAdmin

Delete Memory (Agent memory remove)

delete_memory
Destructive

Preview first, 30-day undo. Removes Dayze Agent memories the authenticated user owns: memory_id (or up to 50 memory_ids) from get_memories. Rows get_memories marks source "event" are calendar events (use delete_event). An unknown or other-user id is not found and nothing changes. Dayze chat and get_memories stop using them at once; rebuilds life_state. Reviewed and reversible: first call preview_destructive_change({ tool: "delete_memory", arguments: { memory_id } }), which stores the exact change without applying it. After the user confirms, call delete_memory with preview_id, restore_window_acknowledged: true and request_id; the reviewed preview is applied exactly and a changed record is refused. Returns restore_id; restore_change({ restore_id }) undoes it for 30 days unless the record was edited since. Requires API key or OAuth with context.delete. Share tokens cannot write. ($0.10; API key required)

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
memory_idNoOptional: the memory the preview named (checked against it). A preview of several memory_ids names none.
preview_idYespreview_id returned by preview_destructive_change for this tool.
request_idYesRequired idempotency key; retries return the original receipt.
idempotency_keyNoAlias for request_id.
restore_window_acknowledgedYesMust be true after reviewing the preview and its 30-day restore window.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
accountNoThe Dayze account this data belongs to. With more than one Dayze connection, name it in the answer.
deletedYes
memory_idNoThe memory the preview named; null for a preview of several.
memory_idsYes
preview_idNo
provenanceNoWhich connector wrote the record and which connected Dayze account received it.
restore_idNoPass to restore_change to undo within restore_window_days.
reversibleNo
idempotent_replayNo
life_state_rebuiltYes
restore_expires_atNo
restore_window_daysNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • addedOutput schema / properties / account
      Added value: +{
+  "additionalProperties": false,
+  "description": "The Dayze account this data belongs to. With more than one Dayze connection, name it in the answer.",
+  "properties": {
+    "display_name": {
+      "description": "Account display name.",
+      "type": [
+        "string",
+        "null"
+      ]
+    },
+    "handle": {
+      "description": "Account handle, e.g. @goh.",
+      "type": [
+        "string",
+        "null"
+      ]
+    },
+    "note": {
+      "description": "How to disclose the account to the user.",
+      "type": "string"
+    },
+    "qa_fixture": {
+      "description": "True for a known QA test account: its records are fixture data, not a real life.",
+      "type": "boolean"
+    }
+  },
+  "required": [
+    "handle",
+    "display_name",
+    "qa_fixture",
+    "note"
+  ],
+  "type": "object"
+}
    • addedOutput schema / properties / provenance
      Added value: +{
+  "additionalProperties": false,
+  "description": "Which connector wrote the record and which connected Dayze account received it.",
+  "properties": {
+    "account": {
+      "additionalProperties": false,
+      "description": "The Dayze account this data belongs to. With more than one Dayze connection, name it in the answer.",
+      "properties": {
+        "display_name": {
+          "description": "Account display name.",
+          "type": [
+            "string",
+            "null"
+          ]
+        },
+        "handle": {
+          "description": "Account handle, e.g. @goh.",
+          "type": [
+            "string",
+            "null"
+          ]
+        },
+        "note": {
+          "description": "How to disclose the account to the user.",
+          "type": "string"
+        },
+        "qa_fixture": {
+          "description": "True for a known QA test account: its records are fixture data, not a real life.",
+          "type": "boolean"
+        }
+      },
+      "required": [
+        "handle",
+        "display_name",
+        "qa_fixture",
+        "note"
+      ],
+      "type": "object"
+    },
+    "channel": {
+      "description": "Always mcp_connector.",
+      "type": "string"
+    },
+    "connector": {
+      "additionalProperties": false,
+      "properties": {
+        "kind": {
+          "description": "oauth or api_key.",
+          "type": "string"
+        },
+        "name": {
+          "description": "The connected app or API-key label shown to the account owner.",
+          "type": "string"
+        }
+      },
+      "required": [
+        "kind",
+        "name"
+      ],
+      "type": "object"
+    },
+    "source": {
+      "description": "chatgpt for ChatGPT OAuth writes; mcp for another OAuth app or API key.",
+      "type": "string"
+    }
+  },
+  "required": [
+    "source",
+    "channel",
+    "connector",
+    "account"
+  ],
+  "type": "object"
+}
  2. Added

TDQS

A4.8/5.0
Behavior5/5

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

Annotations only declare destructiveHint=true; the description goes far beyond by disclosing the 30-day restore window, that a changed record is refused, downstream effects (chat and get_memories stop using them at once, life_state rebuilds), the auth requirement (API key or OAuth with context.delete, share tokens cannot write), and idempotent retry behavior. These are exactly the operational traits an agent needs for a destructive tool.

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?

Front-loaded with the key constraint ('Preview first, 30-day undo'), and every sentence carries substantive information. It is dense and occasionally run-on, but there is no filler and no repetition of the schema.

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?

An output schema exists, so return-value detail is unnecessary, and the description still covers the receipt/restore_id concept. Combined with the preview workflow, auth constraints, side effects, and reversibility, nothing an agent needs to invoke this correctly 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?

Schema coverage is 100%, so the baseline is 3, and the description adds workflow meaning: memory_id is optional and checked against the preview, preview_id must come from preview_destructive_change, and restore_window_acknowledged must be true after reviewing the preview. It explains why these parameters exist rather than only restating their types.

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?

States a specific verb and resource ('Removes Dayze Agent memories the authenticated user owns') and scopes ownership explicitly. It distinguishes itself from siblings by routing calendar-event rows to delete_event and by naming get_memories as the source of memory_id. An agent can identify the target object class without opening the schema.

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?

Gives an explicit two-step workflow: call preview_destructive_change first, then call delete_memory with preview_id, restore_window_acknowledged: true and request_id after user confirmation. It names the alternative (delete_event) and the condition that selects it, and states that unknown or other-user ids simply resolve to not-found.

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

All Dayze — Life Context 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