netcafe-memory

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

Annotations already provide destructiveHint=true and readOnlyHint=false, so the bar is lower. The description adds 'you own' as an authorization constraint and reinforces permanence with 'permanently delete,' but does not elaborate on idempotency (despite idempotentHint) or error behavior. This is moderate additional context.

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 a single, front-loaded sentence with an inline example, containing zero wasted words. It is concise and structured for quick scanning.

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?

For a simple single-parameter destructive tool, the description covers what, when, ownership, and provides an example. With an output schema present, no gap is evident; it is contextually complete for its complexity.

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?

Schema coverage is 100% with a full description of 'id'. The description restates 'id from recall' and gives a call example, but does not add new semantic meaning beyond the schema, so baseline 3 is appropriate.

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 states 'Permanently delete one memory you own' with a clear verb (delete) and resource (memory). It distinguishes from siblings like recall and remember by specifying the destructive action and the source of the id.

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?

It provides a clear condition: 'Use when a memory is wrong or sensitive.' It also implies the prerequisite of obtaining the id from recall, but does not explicitly name alternatives or exclusions beyond that, so it's clear context without exhaustive guidance.

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

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

Annotations already cover read-only and non-destructive behavior. The description adds a prerequisite (needs a workspace/key for durable memory) and explains the type of context it retrieves (decisions, fixes, preferences). The example GET request also illustrates the call style. No contradiction with 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 concise and front-loaded with the core purpose. It includes an example and a key prerequisite in two sentences, with no redundant information. Every sentence serves a purpose.

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 the tool's purpose, when to use it, a usage example, and an authentication prerequisite. With full parameter documentation, an output schema, and annotations, the description is sufficiently complete. It also positions the tool well relative to its siblings.

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 schema provides detailed descriptions for all three parameters (limit, query, project), so the description adds little beyond restating that filtering is optional and showing an example that uses 'query.' With 100% schema coverage, the baseline is 3, and the description does not significantly enrich parameter meaning.

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 opens with 'Retrieve previously stored memories,' a specific verb and resource. It also specifies optional filtering by search query and project, and gives a concrete use case (restoring context at the start of work). This clearly distinguishes it from sibling tools 'remember' and 'forget.'

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 says to call it 'at the start of work on a known project' to restore context, which is clear when-to-use guidance. It does not explicitly name alternatives, but the phrasing and sibling names imply that 'remember' is for storing and 'forget' for deleting. The example and note about needing a workspace/key add practical context.

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

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

Annotations only indicate the tool is not read-only, not open-world, not idempotent, and not destructive. The description adds substantial context by disclosing persistence across sessions, per-network limits for anonymous callers, cross-machine sharing with an AllRouter key, and a prohibition on secrets/logs. This goes beyond the annotations and provides meaningful behavioral expectations.

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 a single dense paragraph of roughly 100 words. The first sentence states the core purpose, followed by scoping behavior, a security warning, and a concrete example. Every sentence contributes unique information, and it avoids repetition of schema fields.

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?

Given the tool's moderate complexity (3 parameters, one enum, output schema present), the description covers all essential aspects: what to store, how scope is determined by auth, what not to store, and a usage example. Since an output schema exists, the lack of return-value detail is acceptable. The description is complete for an agent to select and invoke the tool correctly.

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?

Schema coverage is 100% and each parameter already has a clear description: content defines self-contained ≤2000 chars, kind has enum with default, project scopes recall. The description adds an example call but does not materially extend parameter meaning beyond what the schema provides. The baseline of 3 applies.

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 opens with a specific verb and resource: 'Persist a durable memory' and enumerates concrete content types (architecture decision, stable user preference, verified bug fix, important discovery). It also differentiates from siblings by explaining shared versus per-network memory pools, and the example makes the tool's function unmistakable.

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 clearly states when to use the tool by listing appropriate memory types and explicitly warns 'Do not store secrets or raw logs.' It also explains the distinction between anonymous and keyed callers, which helps with selection. However, it does not explicitly name alternative tools (e.g., 'use recall to retrieve'), so it falls short of the explicit alternatives criterion.

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

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

Beyond the readOnlyHint and idempotent annotations, the description adds that the tool is deterministic, calls no model, costs nothing, and never runs out of quota. It also discloses that it returns ready-to-run examples and recipes, providing valuable behavioral context without contradicting 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 three sentences, front-loaded with the core action. Every sentence contributes value: purpose, output type, and usage guidance. There is no fluff or redundancy.

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 tool is simple (one parameter, read-only, output schema provided). The description covers purpose, usage, behavioral traits, and cost/quota characteristics, making it sufficiently complete for an agent to decide when to invoke it.

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 schema already fully documents the 'task' parameter with examples (100% coverage). The description adds the nuance that any language is accepted, which is a small but useful extension beyond the schema, warranting a score above the baseline of 3.

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 clearly states the tool's function: given a plain-language task, it returns the matching tools with example calls and multi-step recipes. It also distinguishes itself from sibling memory tools (forget/recall/remember) by focusing on tool discovery, so there is no ambiguity about what it does.

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 advises to 'Call this FIRST when you are not sure what this server offers', giving clear when-to-use guidance. It also contrasts with 'reading the whole catalogue and guessing', implying it should be preferred over manual exploration.

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