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

With no annotations, the description carries the disclosure burden. It reveals that the tool is adversarial, uses a different model family, and hunts for flaws, edge cases, security issues, and assumptions. This adds meaningful behavioral context beyond a generic 'review' statement, though it does not mention potential costs, rate limits, or output format.

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 with no waste. The first sentence states the core function, the second provides explicit usage guidance and behavioral detail. Very well structured and front-loaded.

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 fully covers purpose, usage, and behavioral nuance for a tool with all-optional parameters and no output schema. It is complete for an agent to decide when to invoke it and what to expect.

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 of the 5 parameters having a clear description. The tool description adds no additional parameter information, so it neither improves nor harms understanding. 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 the tool's purpose with a specific verb ('Get'), a resource ('adversarial second opinion'), and a distinguishing detail (from Gemini Pro). It names concrete use cases (plan critiques, design reviews, pre-merge code review), which separates it from sibling tools like analyze_files or deep_search.

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 explicit when-to-use guidance: 'ALWAYS use this for plan critiques, design reviews, and pre-merge code review.' This clearly directs the agent for these scenarios and implicitly indicates other tools for other tasks.

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 carries the full burden of behavioral disclosure. It adds a key behavioral trait: 'The files never enter your context — only the answer does,' which is crucial for understanding how the tool operates (offloading analysis to an external CLI). However, it does not mention any potential side effects, prerequisites, or failure modes, which would elevate it to a 5.

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 two sentences long. The first sentence states the core action, and the second provides concrete usage triggers with examples. Every word is purposeful, no filler, and the key information is front-loaded. This is exemplary conciseness.

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 purpose, usage criteria, and a key behavioral trait (context isolation). With a 100% schema-covered parameter set and no output schema, it is nearly complete. It could be slightly enhanced by explicitly stating the answer format (e.g., 'returns a text summary'), but the phrase 'only the answer does' implies a return value. Overall, the tool is well-contextualized.

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%, so each parameter already has a clear description. The tool description adds no additional parameter-level detail (e.g., format, constraints) beyond what the schema provides. Baseline 3 is appropriate when the schema does the heavy lifting.

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 purpose: 'Delegate file analysis to the Antigravity CLI (Gemini) instead of reading files yourself.' This uses a specific verb ('delegate'), a resource ('file analysis'), and explicitly distinguishes itself from direct file reading, which aligns with the sibling tools being web/analysis-oriented. It is unambiguous and unique.

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?

Explicit usage criteria are provided: 'USE THIS whenever a file is large (>200 lines) or the task spans more than 3 files: logs, database dumps, generated code, cross-file reviews, comparisons.' It also implies when not to use it ('instead of reading files yourself'), giving clear when/when-not guidance. This is a model of usage guideline clarity.

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 carries full burden. It adds context about saving the LLM's context window and delegates to the Antigravity CLI, implying subprocess execution, but does not explicitly state read-only behavior or output format. Some behavioral details like whether it modifies the repo remain implicit.

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 two concise sentences front-loaded with the core purpose, followed by concrete examples and a directive. Every part adds value 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?

For a tool with three parameters and no output schema, the description covers the purpose and intended use but not the return format or side effects. Without annotations, some behavioral information is absent, though the description is adequate for a code-search tool. A note about output structure would improve completeness.

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% for all three parameters, so the baseline is 3. The description's mention of 'when did X change' and 'where is Y used' loosely connects to the query parameter but adds no additional semantics beyond the schema. No credit beyond baseline.

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 specific verbs and resources: 'Delegate codebase archaeology', 'git log/diff/blame spelunking', 'wide greps'. It clearly distinguishes this from sibling tools by focusing on codebase history and search, which is a distinct purpose.

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 directs when to use: 'USE THIS instead of running many search commands yourself — it saves your context.' This provides a clear alternative and contextual trigger, though it doesn't name sibling tools, the directive is specific and actionable.

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 transparency burden. It usefully discloses that agy has full tool access (shell, file reads, web) in the cwd, which signals potential side effects. However, it doesn't mention output format, execution duration, error behavior, or whether delegated runs can be destructive beyond the generic 'full tool access' warning.

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 deliver purpose, usage scope, and a critical behavioral warning. No wasted words, front-loaded with the core action.

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 delegation tool with no output schema and no annotations, the description covers the essential invocation context: what to send, where it runs, and what capabilities the delegated agent has. It lacks a note on response handling or long-running task behavior, but is adequate for an agent to 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?

All three parameters are documented with 100% schema coverage, so the structural baseline applies. The description adds no extra semantics beyond hinting that the prompt is the complete task and that cwd defines the execution context, which the schema already communicates.

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 clearly states the action (delegating to the Antigravity CLI), the resource (Antigravity CLI/agy), and the scope (heavy tasks that don't fit other tools). It explicitly distinguishes from sibling tools by framing this as the fallback for workloads outside their coverage.

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 'for heavy tasks that don't fit the other tools' gives clear when-to-use context and implies the sibling tools are preferred for lighter or more specific jobs. It doesn't enumerate exclusions or name siblings explicitly, but the guidance is clear enough for an agent to route correctly.

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?

No annotations are provided, so the description carries the behavioral disclosure burden. It usefully reveals that the server retains full prior context and that the client does not resend history, but it does not mention potential side effects, required permissions, or what the response contains.

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 two short, front-loaded sentences. Every clause either clarifies the tool's purpose, explains when to use it, or reveals stateful behavior, with no unnecessary 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?

For a simple tool with four parameters and full schema coverage, the description covers the purpose, usage context, and the key stateful behavior. It omits the return shape, but that is not likely to be a blocker for a conversational follow-up tool.

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%, so the baseline is 3, but the description adds meaningful parameter context by noting that session_id is 'returned by every other tool.' This helps the agent know exactly where to source a required value beyond what the schema states.

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 that the tool continues a previous Antigravity session by session_id and explicitly frames it for follow-up questions about a prior delegation. This distinguishes it from sibling tools like delegate, which would be used for starting something new.

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 an explicit 'USE THIS' directive for follow-up questions about a prior delegation, and explains why: the full prior context is already on the server side so nothing needs to be resent. It does not explicitly name exclusions like 'use delegate for a new task,' but the intended usage is clear.

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 burden. It transparently reveals that the tool delegates to the Antigravity CLI with web access, implying an external network callfox. It doesn't mention side effects or errors, but for a lookup tool it's adequate. It avoids contradicting any annotations since none exist.

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 two sentences, with the first sentence stating the purpose and the second giving a clear condition for use. It is front-loaded and contains no 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 lookup tool with no output schema, the description provides sufficient context: it explains what the tool does times, what kinds of information it can retrieve, and when to use it. However, it doesn't mention any output format or limitations, which would be helpful but not critical for such a tool.

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 describes the parameters clearly (100% coverage), so the baseline is 3. The description adds context by listing example content for the query (library docs, API references, etc.), enriching the semantics of the 'query' parameter beyond the schema's generic description. Thus, a 4 is warranted.

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 specifies a web/documentation lookup delegation to the Antigravity CLI, listing concrete use cases like library docs, API references, and error messages. This distinguishes it from sibling tools by focusing on external information retrieval, though it doesn't explicitly name alternatives.

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?

Explicitly states 'USE THIS when you need information you don't have or that may be newer than your training data.' This gives a clear condition for use, effectively guiding the agent on when to select this tool.

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