Skip to main content
Glama

memxus

by com.memxus

Forget memory

forget
Destructive

Permanently delete one memory by UUID. When to use: user asks to remove outdated or incorrect context, or to free plan storage. When NOT: fix content → update (mode=replace); find the ID first → list_memories or recall. Requires delete OAuth scope. Non-idempotent: deleting the same memory_id twice fails. Errors: Memory not found, Not authorized to delete this memory. Side effects: removes the memory row and vector embedding with no recovery; invalidates plan cache. The target workspace is always the one the memory itself belongs to (echoed in resolved_workspace); optionally pass workspace: as a safety confirmation — the call fails if the memory is not actually in that workspace.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
memory_idYesUUID of the memory to delete. Copy from the id field in list_memories or recall. Must be deletable by the authenticated user.
workspaceNoTo operate on a team workspace, pass its exact name, slug, or ID (e.g. "Acme"). Omit — or pass "personal" — for your personal memory (default). Every response echoes resolved_workspace so you can confirm where the operation actually happened. Call list_collections when unsure of the exact workspace name.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
deletedYesTrue when deletion succeeded.
messageYesHuman-readable confirmation (same as content text).
memory_idYesUUID of the deleted memory.
resolved_workspaceNoThe workspace this call actually operated on (defense against writing to the wrong team by typo or name collision). id=null means Personal.

TDQS

A4.9/5.0
Behavior5/5

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

The description adds significant context beyond the annotations: it states the tool requires 'delete OAuth scope', is 'non-idempotent' (matching idempotentHint=false), and details side effects ('removes the memory row and vector embedding with no recovery; invalidates plan cache') and errors ('Memory not found, Not authorized'). This fully compensates for the lack of annotation detail.

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?

The description is well-structured with clear sections (purpose, usage, scope, idempotency, errors, side effects, workspace details). While it is somewhat lengthy, every sentence adds value and there is no redundant information. A slight trimming could make it more concise, but it remains highly effective.

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?

Given the tool's complexity (destructive, non-idempotent, auth-scoped, workspace-dependent), the description covers all necessary aspects: when to use, prerequisites, idempotency, side effects, error handling, and workspace safety confirmation. It references the output schema implicitly (via 'resolved_workspace') and is fully self-contained.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema has 100% coverage, and the description adds valuable meaning: for 'memory_id', it advises to 'Copy from the id field in list_memories or recall' and specifies 'Must be deletable by the authenticated user.' For 'workspace', it explains default behavior, personal vs. team, and how to confirm via 'resolved_workspace', plus a reference to 'list_collections'. This goes well beyond the bare 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 description clearly states the tool's function: 'Permanently delete one memory by UUID.' It uses a specific verb ('delete') and resource ('memory') with an identifier, and distinguishes from sibling tools like 'update' (mode=replace) and 'list_memories' or 'recall' (to find the ID).

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?

The description explicitly states when to use the tool ('user asks to remove outdated or incorrect context, or to free plan storage') and when not to ('fix content → update'; 'find the ID first → list_memories or recall'). It also mentions alternatives like 'update' and 'list_memories/recall', providing clear guidance.

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

All memxus toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A4.4/5.0
Disambiguation5/5

Each tool has a clear, distinct purpose: forget deletes, remember saves, update modifies, recall searches, list_memories lists, get_memory retrieves full details, get_context builds a context block, list_collections lists folders, memory_stats shows counts. There is no functional overlap that would confuse an agent.

Naming Consistency3/5

Names mix conventions: forget, recall, remember, update are single verbs; get_context, get_memory, list_collections, list_memories follow verb_noun; memory_stats is noun_noun. This inconsistency can make it harder for an agent to predict tool names.

Tool Count5/5

With 9 tools, the server is well-scoped for a memory management system. Each tool addresses a core operation (CRUD, search, stats, context building) without unnecessary redundancy. The count is appropriate for the domain.

Completeness4/5

The surface covers create (remember), read (list_memories, get_memory, recall, get_context), update (update), delete (forget), plus metadata tools (list_collections, memory_stats). Missing explicit filtering by tags or bulk operations, but these are minor gaps given the semantic search capability.

Resources