Semantic search over your stored memories — finds relevant notes by meaning, not just keywords (e.g. a query about "deployment" surfaces a memory that says "ships on Cloudflare Workers"). Call this at the start of a task, or whenever you might have prior context about the user, project, or topic.
| Name | Required | Description | Default |
|---|---|---|---|
| limit | No | Max results (default 10). | |
| query | No | What to look for, in natural language. Leave empty to get the most recent memories. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description explains the semantic matching behavior with an example, which goes beyond the tool's name. However, it does not explicitly state whether the operation is read-only, nor does it describe any potential side effects or return format, which would be expected given the absence of annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences: the first defines the tool's core function, the second gives usage guidance. Every sentence carries information without excessive length.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description covers purpose, usage, and gives a concrete example, which is adequate for a simple search tool with no output schema. It could be improved by explicitly mentioning the return format (e.g., list of matching memories) but the current level is sufficient for an agent to understand and invoke the tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already documents both parameters (query and limit) with descriptions, so schema coverage is 100%. The description adds semantic value by illustrating how query should be phrased with the 'deployment' example, but doesn't add details about limit beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description defines recall as 'Semantic search over your stored memories' and contrasts it with keyword search, using a concrete example. This clearly distinguishes it from sibling tools (remember, forget, list_recent) and identifies the specific resource (memories).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly states when to call it: 'at the start of a task, or whenever you might have prior context about the user, project, or topic.' It provides clear usage context but does not explicitly mention when not to use it or alternative tools like list_recent for recent memories.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Each tool has a clearly distinct purpose: remember stores, recall searches semantically, list_recent lists by recency, and forget deletes by ID. There is no overlap in functionality, so an agent can easily select the right tool.
Most tool names are concise imperative verbs (forget, recall, remember), while list_recent follows a verb_noun pattern. This minor inconsistency is still readable and predictable, as all names use lowercase with no mixed styles.
With 4 tools, the server is well-scoped for a memory system. Each tool covers a core operation (create, search, list, delete) without unnecessary bloat, making the number appropriate.
The memory lifecycle is fully covered: remember (create), recall/list_recent (read/search), and forget (delete). There is no missing operation that would hinder an agent from managing memories effectively.