kestrel

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

With no annotations, the description carries the full burden. It discloses a key behavioral trait: 'Updates the baseline to the new content.' However, it omits other relevant behaviors such as error handling (e.g., invalid watch_id) and does not state whether the update is reversible or idempotent. This is partial disclosure.

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, with two sentences. The first sentence front-loads the core action and outcome; the second provides usage guidance. No redundant words or 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?

The tool is simple with two parameters, but there is no output schema. The description vaguely says it 'report[s] whether it changed' without specifying the exact format (e.g., boolean, status string, or structured list). It also does not cover error scenarios. For a simple tool this is a moderate gap, but not severe.

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 only 50% (token has a description, watch_id does not). The description adds no parameter-specific information and does not compensate for the missing watch_id documentation. An agent may not know what watch_id refers to beyond inference from the tool name.

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 action: 'Re-fetch a watched page and report whether it changed since the last check, with a short list of added/removed content lines.' This uses a specific verb (re-fetch/report) and resource (watched page), and clearly differentiates from siblings like create_watch and delete_watch.

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 on when to use: 'Call this every time you want the latest status (e.g. each run/heartbeat).' This provides clear context but does not explicitly mention alternatives or exclusions, though the sibling set makes the distinction obvious.

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?

With no annotations, the description carries the full burden of behavioral disclosure. It explains that create_watch captures a baseline snapshot immediately, returns a watch_id and a token, and that the token is required for later check/manage operations. This goes beyond the base function and equips the agent with the non-obvious account-grouping behavior.

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?

Three dense sentences, front-loaded with the core action and immediately followed by the crucial output/token guidance. No filler words; every sentence contributes necessary operational detail.

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 is strong for a creation tool: it names the returned identifiers, warns the agent to save the token, and explains the token reuse case. Since there is no output schema or annotations, a bit more detail about what 'monitoring' entails (e.g., polling behavior or failure modes) could improve completeness, but nothing essential for a first call is missing.

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?

All three parameters have schema descriptions (100% coverage), so the baseline is 3. The description adds genuine value beyond the schema by explaining the token's role in grouping watches and persistence: 'Pass an existing token to add more watches under the same account.'

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?

Description uses a specific verb and resource: 'Start monitoring a web page (or JSON/text URL) for changes.' This clearly distinguishes create_watch from its siblings (check_watch, delete_watch, list_watches) by describing the creation lifecycle, including capturing a baseline snapshot.

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 signals this is the setup tool: you start monitoring and receive credentials for later operations. It doesn't explicitly list when not to use it or name alternatives, but the lifecycle context and the instruction to pass an existing token for grouped watches provide adequate usage 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?

With no annotations provided, the description must disclose behavioral traits. It states that the tool stops monitoring and deletes the watch, which conveys the primary destructive action. However, it does not mention any side effects, irreversibility, permission requirements, or behavior when the watch does not exist. The description is minimally informative but not contradictory.

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, concise sentence that is front-loaded with the primary verb. There is no redundant information, and it is appropriately sized for the tool's simplicity.

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 absence of annotations, output schema, and parameter explanations, the description is incomplete. It lacks details on the token's purpose, expected input formats, and any success or failure behaviors. The tool is simple, but the description still leaves the agent without enough information to confidently invoke it 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 description coverage is 0%, and the description does not explain the meaning or purpose of the parameters (watch_id and token). It only names the fields without adding context, forcing the agent to infer that watch_id identifies the watch and token is likely an authentication credential. This is insufficient for a tool with no schema descriptions.

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 a specific verb (delete) and resource (watch), and the action of stopping monitoring. It distinguishes from siblings like check_watch, create_watch, and list_watches by indicating a destructive action. This is 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 provides no guidance on when to use this tool versus its siblings (check_watch, create_watch, list_watches). It does not mention any conditions, prerequisites, or alternative scenarios. The agent is left to infer that deletion is appropriate when a watch is no longer needed, but this is not explicitly stated.

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?

With no annotations, the description carries the full burden of behavioral disclosure. It adds value by mentioning the output includes 'check/change counts,' but it does not disclose pagination, result ordering, empty-list behavior, or whether the list is filtered in any way beyond the token scope.

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?

One sentence with no filler; the core action is front-loaded and the output detail is appended cleanly. Every word contributes to understanding the tool.

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 one-parameter list tool, the description states the resource, scope, and an output detail, which is adequate. However, because there is no output schema, the description does not clarify the exact response format, whether the list is paginated, or how counts are structured, leaving some uncertainty for an agent.

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 zero descriptions for the token parameter, so the description's phrase 'under your token' adds the meaning that the token is an authentication/scope identifier. This is useful but minimal; it does not explain the token's format, how to obtain it, or any validation requirements.

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 a specific action: 'List all watches under your token' with an explicit output detail of 'per-watch check/change counts.' This clearly identifies the tool as a read operation for enumerating watches, distinguishing it from siblings that check, create, or delete individual watches.

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 phrase 'List all watches' implies the tool is for enumeration, and the sibling names (check_watch, create_watch, delete_watch) make the contrast obvious. However, the description does not explicitly state when to use this tool versus those alternatives, nor provide any exclusionary guidance such as 'use check_watch for a single watch.'

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