Server Quality Checklist

Tool Scores

• read_annotationsA
  3.8/5.0
  Behavior2/5

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

  No annotations are provided, so the description must disclose behavioral traits on its own. It describes the data contents in detail, but uses the ambiguous phrase 'consume user feedback' without clarifying whether reading is destructive or marks annotations as no longer pending. Since 'clear_annotations' exists, reading is likely non-destructive, but the description fails to confirm this, creating uncertainty about side effects.

  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 two sentences with no wasted words. The first sentence is dense, listing all the fields in a compact manner, while the second gives the usage timing. It could be structured with a list for readability, but it remains appropriately concise.

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

  Completeness3/5

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

  With no output schema, the description thoroughly enumerates the return structure (CSS selector, fallback chain, comment, type, color, viewport position, metadata), which provides strong coverage. However, it leaves the lifecycle of 'pending' annotations ambiguous and doesn't clarify whether reading has side effects. Given the low tool complexity, this omission prevents a higher score.

  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?

  There are zero parameters, making schema coverage 100%. The description adds context about what the read operation returns, which is the relevant semantic contribution when no parameters exist. Baseline for 0 params is 4.

  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 opens with 'Read all pending visual annotation feedback' – a specific verb+resource pair. It then enumerates the annotation fields (selector, fallback chain, comment, type, color, viewport position, element metadata), clearly distinguishing it from the sibling 'clear_annotations' which focuses on removal.

  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?

  The final sentence provides an explicit usage trigger: 'Use this to consume user feedback after they press Submit in the overlay.' It does not explicitly state when not to use it, but with only one sibling (clear_annotations) the contrast is implied, and the timing criterion is clear.

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

• clear_annotationsA
  4.2/5.0
  Behavior3/5

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

  No annotations are provided, so the description carries the burden. It discloses the destructive nature ('Clear all') and adds workflow context, but does not mention irreversibility, scope (e.g., global vs workspace), or permissions. For a zero-parameter tool this is adequate but not thorough.

  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?

  Two short sentences with the action stated first. No filler. Every word earns its place.

  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?

  Given the tool's simplicity (0 params, no output schema), the description sufficiently covers its purpose and usage timing. It could mention that clearing is irreversible or affects all annotations, but those are not necessary for basic invocation.

  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 has zero parameters, so the schema is trivially 100% covered. The description adds no parameter-specific details, but none are needed; baseline of 4 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 uses a specific verb ('Clear') and resource ('all stored annotations'), making the action unmistakable. It also distinguishes itself from the sibling tool 'read_annotations' by explicitly focusing on clearing rather than reading.

  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?

  The description clearly states when to use the tool ('Call after processing feedback to prepare for the next annotation round'). While it doesn't discuss alternatives or when not to use it, the context is clear and the single sibling tool is distinct enough.

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