Server Quality Checklist

Tool Scores

• memory_addA
  3.6/5.0
  Behavior3/5

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

  With no annotations provided, the description carries the full behavioral disclosure burden. It does disclose the most important behavior: memory persists across sessions and processes. But it does not explain what happens if the same fact is added multiple times, whether overwrites/corrections replace existing memory, or what the tool returns or fails on.

  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 relatively tight and front-loaded with the core action. The list of example memory types is somewhat redundant with the phrase 'worth carrying forward,' but each sentence still contributes to deciding when and what to invoke.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a simple write/mutation tool with three parameters and no output schema, the description gives sufficient operational context: what to store, when to call, and the persistence guarantee. It lacks minor detail about edge cases or return behavior, but is likely adequate for an agent 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 coverage is 100%, so the schema already defines content, tags, and importance. The description adds mild guidance about what kind of content should be stored (preferences, corrections, decisions), but it does not add meaningful semantic detail beyond the schema's parameter descriptions.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool's action ('Store a durable fact, preference, decision, or correction') and its outcome ('FUTURE sessions remember it'). It is distinguishably an add/write operation compared to siblings like memory_search and memory_forget, though it never names these siblings explicitly.

  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?

  Provides clear triggering context: 'Call this whenever the user states something worth carrying forward' and lists examples like preferences, facts, and corrections. However, it does not state when not to use the tool or explicitly point to alternative tools like memory_search or memory_forget.

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

• memory_stateA
  4.5/5.0
  Behavior4/5

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

  With no annotations, the description carries the full transparency burden, and it does well: it discloses that the tool reads the store directly, requires no network or embedding, and notes the selfFacets-is-always-0 quirk with its root cause (no consolidation pass). This adds meaningful behavioral information an agent would not infer from the name or empty schema.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is this concise and well-structured: the main sentence states the function, the second sentence adds differentiating behavior (no embedding/network), the third gives usage guidance, and the parenthetical discloses a critical invariant. No word is wasted, and the most important information is front-loaded.

  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?

  For a read-only, zero-parameter tool with no output schema, the description is fully sufficient: it explains what the tool returns, when to call it, what side effects it does not have, and the only significant edge case. The agent has everything needed to invoke it correctly.

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

  Parameters4/5

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

  The tool accepts zero parameters, so the description does not need to explain parameter behavior. It instead clarifies what the output contains (counts and recent facts), which is the appropriate compensating information. The baseline of 4 for zero-parameter tools applies.

  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 states a specific verb ('Summarize') and a precise resource ('what is currently in memory'), with output detail ('counts plus the most recently stored facts'). It also distinguishes itself from search-like siblings by noting it reads the store directly without embedding calls or network access, so an agent can tell this from memory_search at a glance.

  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?

  It gives explicit usage context: 'Use to orient yourself at session start, or when the user asks what you remember.' This is clear and actionable. However, it does not explicitly state when to prefer a sibling alternative such as memory_search for a specific fact lookup, so it stops one step short of full exclusion guidance.

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

• memory_searchA
  4.7/5.0
  Behavior4/5

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

  With no annotations present, the description carries the full burden. It discloses the retrieval mechanism ('MMR retrieval, not a text match') and states that it returns relevant memories, which is meaningful behavioral context beyond the simple verb 'search'. However, it does not explicitly say that the operation has no side effects or whether it reads from a persistent store, though this is strongly implied.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is compact and front-loaded with the core purpose. Each sentence contributes: when to call, why it differs from a simple keyword search, and a fallback policy. There is no redundant or filler content.

  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 a single parameter and no output schema, the description is complete enough. It tells the agent when to use the tool, what retrieval algorithm to expect, and what the result will be ('relevant memories'). An agent can confidently invoke it without further clarification.

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

  Parameters4/5

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

  Schema coverage is 100% and the query parameter is simply described as 'What to recall.' The description adds useful semantic guidance: the query is not a literal text match, so conceptual or paraphrased queries are appropriate, and it suggests what kind of content to search for (past context, preferences, project history).

  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 identifies a specific action ('recall') and resource ('what you already know'), and adds a distinguishing mechanism: 'MMR retrieval, not a text match.' This is sufficiently specific and distinct from sibling tools like memory_add and memory_forget.

  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?

  It provides explicit when-to-use guidance: 'Call this at the START of a session, and whenever the user refers to past context, their preferences, or the project history.' It also gives a default heuristic: 'If unsure whether you know something, search first.' This leaves little ambiguity about when to invoke it.

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

• memory_forgetA
  4.9/5.0
  Behavior5/5

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

  Since no annotations are provided, the description fully carries the behavioral transparency burden. It clearly discloses the destructive, irreversible nature of the operation, the case-insensitive substring matching behavior, and the broad-scope risk of short queries. This level of warning is critical for a destructive delete tool.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is compact and dense, front-loading the critical warning 'DESTRUCTIVE' before any other detail. Every sentence adds operational value, including when to use the tool, how to phrase the query, and what to check beforehand.

  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 destructive nature, the low parameter count, and the absence of an output schema, the description covers all essential aspects: action, matching semantics, irreversibility, risk, usage conditions, and a pre-check alternative. An agent is sufficiently prepared to avoid dangerous misuse.

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

  Parameters4/5

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

  The schema already documents the query parameter completely (100% coverage), but the description adds important semantic detail: the query is matched as a case-insensitive substring against memory content, not as a semantic match or single-item identifier. This extra context meaningfully improves the agent's understanding of the parameter.

  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 action: deletes every memory whose content contains a case-insensitive substring, and explicitly distinguishes this from semantic or single-item delete. This differentiates the tool from siblings like memory_search and memory_add without ambiguity.

  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 provides explicit when-to-use guidance: only after user correction of a stored fact or explicit request to forget. It also warns of broad deletion risk, recommends using a long distinctive phrase, and suggests checking memory_state first when uncertain, giving the agent concrete decision-making guidance.

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