@saihm/mcp-server-pro

Official

by SAIHM-Admin

Overview Schema Source Related Servers Discussions Admin

TypeScript

Remote

Forget (GDPR erasure)

saihm_forget

DestructiveIdempotent

Cryptographically erase one memory by its cell id (GDPR Art. 17). Destroying the wrapped key leaves the content unreadable to everyone, the operator included. Irreversible: use it when erasure is the intent, not to tidy a working set.

Instructions

Input Schema

TableJSON Schema

Name	Required	Description
`id`	No	Memory cell id (hex) to erase. Required unless warmStart is "delete".
`token`	No	With warmStart: 16 to 32 characters from a-z and 0-9, chosen at random for this call. Blank otherwise.
`warmStart`	No	"delete" destroys the snapshot named by snapshotId instead of a memory. Only snapshotId and token go with it. Leave it blank to erase a memory by id.
`snapshotId`	No	With warmStart "delete": the snapshot to destroy. Blank otherwise.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`at`	No
`epoch`	No
`error`	No
`reason`	No
`timing`	No
`deleted`	No
`complete`	No
`sharesPurged`	No
`destructionAnchor`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed6 schema fields changedv0.14.1Oct 14, 2026

changedInput schema / properties / snapshotId / description

Previous value: -"With warmStart \"delete\": the snapshot to destroy."New value: +"With warmStart \"delete\": the snapshot to destroy. Blank otherwise."

changedInput schema / properties / snapshotId / pattern

Previous value: -"^[0-9a-f]{64}$"New value: +"^$|^[0-9a-f]{64}$"

changedInput schema / properties / token / description

Previous value: -"With warmStart: 16 to 32 characters from a-z and 0-9, chosen at random for this call."New value: +"With warmStart: 16 to 32 characters from a-z and 0-9, chosen at random for this call. Blank otherwise."

changedInput schema / properties / token / pattern

Previous value: -"^[a-z0-9]{16,32}$"New value: +"^$|^[a-z0-9]{16,32}$"

changedInput schema / properties / warmStart / description

Previous value: -"\"delete\" destroys the snapshot named by snapshotId instead of a memory. Only snapshotId and token go with it."New value: +"\"delete\" destroys the snapshot named by snapshotId instead of a memory. Only snapshotId and token go with it. Leave it blank to erase a memory by id."

changedInput schema / properties / warmStart / enum

Previous value: -[
-  "delete"
-]New value: +[
+  "",
+  "delete"
+]

Changed6 schema fields changedv0.13.0Oct 7, 2026

changedInput schema / properties / id / description

Previous value: -"Memory cell id (hex) to erase"New value: +"Memory cell id (hex) to erase. Required unless warmStart is \"delete\"."

addedInput schema / properties / snapshotId

Added value: +{
+  "description": "With warmStart \"delete\": the snapshot to destroy.",
+  "pattern": "^[0-9a-f]{64}$",
+  "type": "string"
+}

addedInput schema / properties / token

Added value: +{
+  "description": "With warmStart: 16 to 32 characters from a-z and 0-9, chosen at random for this call.",
+  "pattern": "^[a-z0-9]{16,32}$",
+  "type": "string"
+}

addedInput schema / properties / warmStart

Added value: +{
+  "description": "\"delete\" destroys the snapshot named by snapshotId instead of a memory. Only snapshotId and token go with it.",
+  "enum": [
+    "delete"
+  ],
+  "type": "string"
+}

removedInput schema / required
```
Removed value: -[
-  "id"
-]
```

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "additionalProperties": false,
+  "properties": {
+    "at": {
+      "type": [
+        "string",
+        "null"
+      ]
+    },
+    "complete": {
+      "type": [
+        "boolean",
+        "null"
+      ]
+    },
+    "deleted": {
+      "const": true,
+      "type": "boolean"
+    },
+    "destructionAnchor": {
+      "type": "string"
+    },
+    "epoch": {
+      "type": "string"
+    },
+    "error": {
+      "additionalProperties": false,
+      "properties": {
+        "code": {
+          "type": "string"
+        },
+        "status": {
+          "type": "number"
+        }
+      },
+      "required": [
+        "code",
+        "status"
+      ],
+      "type": "object"
+    },
+    "reason": {
+      "enum": [
+        "not-found",
+        "not-attested"
+      ],
+      "type": "string"
+    },
+    "sharesPurged": {
+      "type": [
+        "number",
+        "null"
+      ]
+    },
+    "timing": {
+      "additionalProperties": false,
+      "properties": {
+        "admission": {
+          "anyOf": [
+            {
+              "enum": [
+                "reserved",
+                "shared"
+              ],
+              "type": "string"
+            },
+            {
+              "type": "null"
+            }
+          ]
+        },
+        "serverMs": {
+          "type": [
+            "number",
+            "null"
+          ]
+        },
+        "totalMs": {
+          "type": "number"
+        }
+      },
+      "required": [
+        "serverMs",
+        "totalMs",
+        "admission"
+      ],
+      "type": "object"
+    }
+  },
+  "type": "object"
+}

First observedv0.2.1Jul 29, 2026

TDQS

A4.4/5.0

Behavior5/5

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

Annotations already declare destructiveHint=true and idempotentHint=true, but the description adds real value beyond them: the mechanism (destroying the wrapped key) and the consequence (unreadable to everyone, operator included). That is exactly the kind of context an agent needs before an irreversible call.

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 tight sentences with the destructive intent and irreversibility front-loaded before the mode variant. Slight redundancy with the schema on the warmStart branch and the trailing "its key first" is a touch cryptic, keeping it from a 5.

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 values need no explanation, and annotations carry the safety profile. The description covers mechanism, consequence, reversibility, and the dual memory/snapshot mode, leaving nothing an agent needs to call it correctly.

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 description coverage is 100%, so the baseline is 3, and the schema already documents the warmStart="delete" branch, id format, and token pattern in more detail than the description. The description restates the mode switch but adds no syntax or format detail beyond the 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?

States a specific verb+resource ("cryptographically erase one memory by its cell id") with the GDPR framing that pins the intent. An agent can distinguish it from siblings like saihm_revoke_share or saihm_remember without opening a schema.

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?

Explicit when-to-use and when-not ("use it when erasure is the intent, not to tidy a working set"), plus a clear branch for the snapshot mode. It does not name a concrete alternative tool for unsharing/tidying, so it stops short of full routing guidance.

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

Deploy Server