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

With no annotations, the description carries the behavioral burden. It reveals the tool uses a different model family, hunts for flaws, edge cases, security issues, and unstated assumptions, which gives a clear sense of its adversarial nature. It doesn't detail side effects or performance, but the core behavior is well disclosed.

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, no fluff. The first sentence defines the tool; the second gives explicit usage directives. Every word earns its place, and the key information is 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?

Given the tool's moderate complexity (5 optional params, no output schema), the description provides sufficient context for selection and invocation. It covers what the tool does, when to use it, and the schema handles parameters. It doesn't explain return format, but that's not critical for this review-style 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 input schema has 100% coverage with descriptive parameter docs. The tool description itself does not add parameter-level semantics, so the baseline of 3 applies. The schema already explains each parameter adequately.

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 provides an adversarial second opinion from a different model family (Gemini Pro), and explicitly lists concrete use cases (plan critiques, design reviews, pre-merge code review). This distinguishes it from sibling tools like analyze_files or deep_search, which serve different purposes.

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 'ALWAYS use this for' the listed review scenarios, giving clear when-to-use guidance. It does not explicitly mention when not to use it or name alternatives, but the strong directive and adversarial framing effectively guide selection.

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. It discloses the key behavioral trait that files never enter the agent's context and only the answer is returned, plus the delegation mechanism (Antigravity CLI/Gemini). It doesn't mention latency, cost, or failure modes, but it does reveal the most important side effect.

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 compact and front-loaded: it states the core action first, then gives crisp usage thresholds and a final context-preservation note. Every sentence earns its place, with no redundant wording.

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 no output schema or annotations, the description covers the essential context: what the tool does, when to use it, and a crucial behavioral trait. It doesn't detail the answer format, but the phrase 'only the answer does' implies a returned answer, which is sufficient for most decisions.

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 covers all parameters at 100%, so the baseline is 3. The description adds some color to the 'files' parameter by listing example file types (logs, database dumps) and to the overall use case, but it doesn't add detailed semantics for 'cwd' or 'model' beyond the 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 identifies the tool as delegating file analysis to the Antigravity CLI (Gemini) instead of reading files directly, and even lists concrete scenarios (logs, database dumps, generated code, cross-file reviews, comparisons). It does not explicitly name sibling tools, but the purpose is unambiguous and distinct from generic file reading.

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?

Provides explicit, actionable guidance: 'USE THIS whenever a file is large (>200 lines) or the task spans more than 3 files' and enumerates specific use cases. This leaves no doubt about when to choose this tool, and implicitly indicates when not to (small/simple cases).

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 burden of behavioral disclosure. It explains that the tool delegates to a CLI and performs various search operations, which implies a read-only nature, but it does not explicitly confirm read-only behavior, mention side effects, output format, or performance implications. The 'saves your context' phrase hints at the tool's execution model but lacks detail.

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, front-loaded with the core purpose and followed by a clear usage directive. Every sentence contributes value, using concrete examples and formatting to make the purpose and benefit immediately clear. No fluff 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?

For a tool with 3 parameters, no output schema, and no annotations, the description provides substantial context: it explains what the tool does, gives usage scenarios, and explains why to use it. It lacks explicit return-value behavior, but given the search/delegation nature, the reader can infer the output. The description is fairly complete for its complexity level, leaving only minor gaps.

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 the schema already documents all three parameters (query, cwd, model) with meaningful descriptions. The tool description adds no additional parameter-related semantics beyond the schema examples, so the baseline of 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 function: 'Delegate codebase archaeology to the Antigravity CLI: git log/diff/blame spelunking, wide greps across a repo, 'when/why did X change', 'where is Y used'.' This is a specific verb+resource with concrete examples, and it distinguishes itself from sibling tools like web_lookup and analyze_files by focusing on deep codebase exploration.

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 usage guidance with 'USE THIS instead of running many search commands yourself — it saves your context.' This indicates when to use the tool (for deep searches to offload work) and implicitly against manual command running. It does not explicitly contrast with sibling tools but does provide clear context and an imperative for use.

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 disclosing behavior. It reveals that this is 'raw' delegation with full tool access (shell, file reads, web) in a given cwd, which is crucial for understanding the tool's power and potential side effects. However, it doesn't mention output behavior or run duration, so not a perfect score.

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 extremely concise at two sentences. It front-loads the core purpose and immediately provides key differentiators and capabilities with zero wasted words.

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 raw delegation tool with no output schema, the description gives enough context: it explains the tool's scope (heavy tasks), capabilities (full tool access), and working directory (cwd). It could mention that prompts should be detailed or that execution may be long, but the essentials are covered.

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 the schema already documents all three parameters. The description adds context about the 'given cwd' but does not add new semantic details beyond the schema. 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: raw delegation to the Antigravity CLI for heavy tasks. It uses a specific verb ('delegation') and resource ('Antigravity CLI'), and distinguishes itself from siblings by explicitly targeting tasks that 'don't fit the other tools'.

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 provides clear context for when to use the tool ('heavy tasks that don't fit the other tools'), but does not explicitly name alternative tools for lighter tasks. The description also highlights agy's full tool access, which hints at appropriate use cases, but could be more explicit about exclusions.

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 bears full burden. It discloses that the prior context is already on the server side, reducing data transfer, but does not mention potential behaviors like session expiration, idempotency, or error conditions. This is adequate but lacks depth.

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 exceptionally concise: two sentences delivering purpose, usage direction, and behavioral hint with zero wasted words. Information is front-loaded and easy to parse.

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 tool's moderate complexity (4 params, no output schema, no annotations), the description covers what the tool does, when to use it, and a key behavioral trait (no resending). It does not explain return format or error scenarios, but for a follow-up tool this is mostly sufficient.

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 schema already describes all four parameters. The description adds no new semantic information beyond what the schema's descriptions provide (e.g., session_id's origin, question's purpose). 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 identifies the tool's purpose: continuing a previous Antigravity session via session_id for follow-up questions. It specifies the resource (session) and action (continue), and the note about not resending context differentiates it from starting new delegations, though explicit sibling differentiation is minimal.

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 use for follow-up questions after a prior delegation, and explains the benefit (no resending context). However, it does not explicitly state when to use alternatives like 'delegate' for new tasks, leaving some inference required.

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 of explaining behavior. It reveals that the lookup is delegated to an external CLI with Gemini web access, implying web freshness. However, it does not disclose potential latency, failure modes, or output format, which would be useful for a tool that depends on an external service.

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, front-loaded with the core action and examples, followed by a clear usage trigger. Every sentence serves a purpose with no filler or repetition.

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 relatively simple lookup tool with full schema coverage and no output schema, the description covers what, when, and the underlying mechanism (external CLI with web access). It is slightly incomplete in not hinting at what the response looks like, but this is not critical for a lookup 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 description coverage is 100%, so the baseline is 3. The description does not add parameter-specific meaning beyond the schema; it only lists example query topics. The schema already documents query, cwd, and model, so the description adds marginal value for parameters.

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 identifies the tool's function: 'Delegate a web/documentation lookup to the Antigravity CLI (Gemini with web access).' It lists specific use cases (library docs, API references, error messages, current versions) and distinguishes it from local-analysis siblings like analyze_files and 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 explicitly states when to use the tool: 'USE THIS when you need information you don't have or that may be newer than your training data.' This is clear guidance, though it does not mention exclusions or directly compare to sibling tools like deep_search.

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