netcafe-memory

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

The description says 'Permanently delete', indicating a destructive, state-changing operation. However, annotations declare readOnlyHint=true and destructiveHint=false, contradicting the description. This is an annotation contradiction, making the behavioral guidance unreliable.

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?

Two concise, front-loaded sentences: the first states the action and object, the second the usage context. No filler 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 description provides core usage context and the tool is simple with an output schema present, so return values need not be explained. However, the contradiction with annotations undermines confidence and leaves the tool's true behavioral profile unclear.

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 description coverage is 100%, with the parameter 'id' already documented as 'Memory id from recall (integer for keyword engine, uuid string for semantic engine)'. The description adds minimal extra 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 uses a specific verb ('delete') and resource ('memory you own'), clearly identifying the action and scope. It also differentiates from sibling tools recall and remember, which retrieve and create memories respectively.

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 use the tool ('Use when a memory is wrong or sensitive'). It does not provide when-not or alternative tool guidance, but the use case is clear and sufficient for a simple tool.

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 declare this as read-only, idempotent, and non-destructive. The description adds behavioral context beyond that: it explains the purpose (restore context: decisions, fixes, preferences) which isn't in annotations. It does not describe return format, but an output schema exists, so the extra context is sufficient.

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?

Two sentences, front-loaded with the core action ('Retrieve previously stored memories'), then a practical usage hint. Every word earns its place; no redundancy or fluff.

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 read-only tool with full annotations, an output schema, and clear parameters, the description fully covers the user's need: it states what it does, when to use it, and why it's valuable. No gaps remain for the given 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 each parameter described ('Optional search terms; omit to list the most recent', 'Optional project filter', 'Max results...'). The description text reinforces query/project filtering but adds no new information beyond what the schema provides, 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 clearly states 'Retrieve previously stored memories' with a specific verb and resource, and distinguishes itself from siblings by being the retrieval counterpart to remember/forget. It also adds scope (optional filtering by query/project), making its purpose unambiguous.

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 gives explicit guidance: 'Call at the start of work on a known project to restore context,' which frames when to use it. However, it does not explicitly mention when not to use it or compare with remember/forget, though the context is clear enough for a read tool.

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?

The description clearly describes a write/persist operation, but the annotations declare readOnlyHint=true. This directly contradicts the tool's stated behavior of storing a memory across sessions. Because the description contradicts the annotations, the score must be 1 regardless of how transparent the prose is.

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?

Two dense sentences carry a wealth of information: persistence, use cases, two tiers of callers, key sharing across machines/agents, and a safety warning. Every clause earns its place with no filler.

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 presence of an output schema and rich annotations (aside from the readOnly contradiction), the description covers the essential behavioral context: durability, pool sizing, key-based sharing, and prohibited content. It stumbles slightly because the readOnlyHint=true annotation could confuse an agent, but the description itself is comprehensive for the tool's 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 each parameter (kind, content, project) already described in the schema itself. The description adds no extra parameter-level meaning beyond the schema; it merely gives examples of content types. This fits the baseline of 3 for complete schema coverage.

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 'Persist a durable memory' – a specific action on a clear resource – and enumerates concrete use cases (architecture decision, user preference, bug fix, discovery). It implicitly contrasts with sibling tools (forget, recall), making the purpose unambiguous.

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 gives clear context on when to use the tool (stable preferences, verified fixes, etc.) and distinguishes between anonymous callers and key-bearing callers with different pool sizes. It also states a firm exclusion ('Do not store secrets or raw logs'), though it doesn't explicitly point to alternatives or when-not cases beyond that.

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