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 only states the high-level purpose and does not explain return types, side effects, permissions, or any limitations. Unlike tools with richer annotations, this provides minimal transparency beyond the obvious.

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 front-loads the purpose without wasted words. However, it omits crucial context that could fit in a few more phrases, so while efficient, it is somewhat under-specified.

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 three parameters, no annotations, and no output schema, the description is severely incomplete. It does not explain what analysis results look like, how to interpret the response, or any constraints. This is insufficient for an agent to use the tool effectively.

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 67% (path and question are described, model is not). The description itself adds no parameter-level semantics and fails to compensate for the undocumented 'model' parameter. The schema descriptions are basic, and the description provides no additional context about parameter usage.

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 performs architectural analysis on files or folders using Antigravity, which distinguishes it from sibling tools like agy_security_audit or agy_refactor. The verb 'perform' and resource 'files or folders' are specific and 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?

There is no guidance on when to use this tool versus alternatives. It does not mention specific use cases, prerequisites, or exclusions, leaving the agent without context for choosing this over other analysis tools like agy_security_audit or agy_notebook_query.

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 for conveying behavioral traits. It does disclose key safety-relevant behaviors: the agent automatically approves modifications and runs commands. However, it does not mention sandboxing, reversibility, execution environment, or potential long-running behavior, leaving important gaps for a high-risk autonomous tool.

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 sentences, front-loaded with the primary purpose and behavioral key points. Every word earns its place, and there is no redundancy or unnecessary 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?

For a complex autonomous agent tool with no output schema and no annotations, the description should explain what the tool returns, how long it may run, and whether the environment is sandboxed. It only gives an overview and omits these critical details, making it incomplete for a tool that can modify and execute commands.

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 description provides no information about any of the three parameters. Schema coverage is 67% (prompt and cwd have descriptions, model does not), but the description does not compensate for the undocumented 'model' parameter or add any context beyond the schema. The parameter semantics are therefore weak.

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 ('Run') and specific resource ('Antigravity agent in fully autonomous mode'), distinguishing it from sibling tools like agy_chat or agy_task by emphasizing the autonomous nature. The scope is well-defined and not a tautology.

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 implies this tool is for autonomous execution ('fully autonomous mode', 'automatically approves modifications') but does not explicitly state when to use it over alternatives or provide any exclusions. There is no mention of when not to use it, so guidance remains implicit rather than explicit.

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 for behavioral transparency. It does disclose a key behavioral trait: the tool is stateful, as it can preserve or resume conversation threads. However, it does not explain side effects such as creating new conversations, returning responses, or any authentication/rate-limit implications, leaving significant gaps.

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 immediately conveys the tool's purpose and key capability (preserving/resuming threads). There is no redundant information or filler, making it highly efficient.

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?

Despite having 5 parameters, no output schema, and no annotations, the description remains very sparse. It fails to explain how to configure the chat (e.g., choosing agent or model), what the response format looks like, or the behavior when no conversation context is provided. This leaves substantial gaps for an agent attempting to invoke the tool 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?

The schema description coverage is 60% (3 of 5 parameters have descriptions), so the baseline is 3. The description adds context by referencing conversation thread management, which aligns with the continueRecent and conversationId parameters, but it does not elaborate on the meaning of agent or model parameters, nor does it clarify the relationship between the state-related 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 as an interactive chat session with Antigravity, with the specific function of preserving or resuming conversation threads. It distinguishes itself from sibling tools by focusing on conversational interaction, although it does not explicitly name any alternative.

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 implies usage for multi-turn conversational interactions by mentioning preserving or resuming conversation threads, but it does not provide explicit guidance on when to use this tool versus alternatives such as agy_autonomous_agent or agy_sandbox_agent. No exclusions or alternative conditions are 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 full burden. It only states the core function but does not disclose any behavioral traits such as whether it is read-only, what the response format is, how the model parameter affects behavior, or any side effects. The description offers no additional context beyond the basic action.

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 one sentence, front-loaded with the primary action, and contains no filler or redundancy. It earns its place without waste.

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 no annotations, no output schema, and incomplete parameter documentation, the description is too sparse. It fails to cover usage context, the optional model parameter, or what the returned explanation looks like. It is minimally viable but has clear 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?

The schema has 50% description coverage (only errorLog is described). The tool description does little to add meaning: it merely restates the errorLog purpose and provides no explanation for the 'model' parameter, leaving it ambiguous. This is a significant gap given the low schema coverage.

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 a specific verb ('Explain') and clearly identifies the resource: compiler, runtime, and stack trace errors. It also states the outcome (actionable fixes), which distinguishes it from sibling tools like agy_analyze or agy_refactor that focus on other tasks.

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 implies usage when an error is encountered (compiler/runtime/stack trace), but it does not explicitly state when to choose this tool over alternatives or mention any exclusions. Sibling tools are not referenced, so guidance is only implicit.

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, but it only states the basic function. It does not mention side effects (e.g., writing files), required permissions, or whether tests are returned or saved, offering minimal transparency.

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 core action. It is appropriately sized with no 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?

Given no annotations, no output schema, and one undocumented parameter, the description is too sparse. It does not convey what the tool returns or creates, any side effects, or usage prerequisites, leaving important contextual 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 provides descriptions for filePath and testFramework (67% coverage), but model lacks a description. The tool description does not clarify model's purpose or how parameters interact, though the mention of 'unit/integration tests' provides some context for testFramework.

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 action (generate), resource (automated unit/integration tests), and target (specified source file). It is specific and distinguishes itself from sibling tools like agy_analyze or agy_refactor by focusing on test generation.

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 alternatives. It does not mention any exclusions, prerequisites, or competing tools, leaving the user to infer appropriate usage.

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?

The verb 'retrieve' implies a read-only operation, but the description does not disclose potential caveats such as exhaustiveness of the list, authentication requirements, or performance characteristics. With no annotations, the description carries the full burden and only partially covers behavioral transparency.

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, clear sentence that directly states the action and object with no unnecessary words. It is front-loaded with the verb and immediately conveys the purpose.

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 zero-parameter list tool, the description adequately conveys what is returned (all subagents and specialized profiles) and the environment (Antigravity CLI). It does not describe the output structure, but given the tool's simplicity, this is not a critical gap.

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 tool has zero parameters, so there is nothing to explain. The schema is empty, making the 100% coverage trivial, and the baseline of 4 for no-parameter tools applies.

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 the specific verb 'retrieve' and clearly identifies the resource as 'the list of all available subagents and specialized profiles' in the Antigravity CLI. This makes the tool's function unambiguous and distinguishes it from siblings like agy_get_models.

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 offers no guidance on when to use this tool versus alternatives. It does not mention scenarios, prerequisites, or alternative tools to consider, leaving the agent to infer usage solely from the name.

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 must carry the behavioral burden. The verb 'Retrieve' implies a read-only operation with no side effects, which is appropriate. However, the description does not disclose any potential caveats such as network access, output format, or whether it returns the full history in one call. Given the simplicity of the operation, this is adequate but not rich.

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 with no redundant words. The main verb and object are front-loaded, and every word contributes meaning.

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 simplicity (zero parameters) and clear purpose, the description is mostly complete. However, there is no output schema, and the description does not specify the return format (e.g., plain text, structured list, paginated). For a changelog retrieval tool, this is a minor gap, but the overall context is 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?

The tool has zero parameters, so the schema is trivially covered. The description adds no parameter-specific details, but none are needed. The baseline for 0 params is 4.

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 the specific verb 'Retrieve' with a clear resource ('official release notes and changelog history of the Antigravity CLI'), making its purpose unambiguous. It is clearly distinct from sibling tools, none of which mention changelog or release notes.

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?

No explicit guidance on when to use this tool vs alternatives is provided. However, the purpose itself implies usage whenever the user needs Antigravity CLI release notes or changelog. There are no competing sibling tools for this resource, so the implied usage is sufficient.

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 only mentions 'Retrieve the list' but does not describe the return format, ordering, pagination, authentication needs, or any side effects, which is insufficient for an agent to know how to handle the response.

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 of 14 words that immediately states the action and object. There is no filler, repetition, or unnecessary 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?

Given the simplicity (zero parameters, no output schema), the description is adequate but minimal. It does not specify the exact structure of the returned list (e.g., array of names vs. objects), which could leave the agent uncertain about how to parse the output. More detail on the return format 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?

The tool has zero parameters and an empty schema, so the baseline is 4. The description correctly does not add parameter information, and no explanation is needed for parameters that do not exist.

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 action ('Retrieve'), the resource ('the list of all available models'), and the context ('supported by the Antigravity CLI'). It is specific and distinguishes this tool from siblings like agy_get_agents, which deal with agents, not models.

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 compared to alternatives, no prerequisites, and no exclusions. It only states what the tool does, leaving the agent to infer usage from context.

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 only says 'Run queries' and 'ingested as context,' offering no information about read-only guarantees, potential side effects like indexing, required permissions, or performance implications.

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, efficient sentence that leads with the action and resource, with the NotebookLM analogy adding value without fluff. Every word earns its place, making it highly concise and readable.

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 annotations, no output schema, and a missing parameter description for 'model', the description is incomplete. It does not explain return format, ingestion behavior, or prerequisites, so an agent may not know what to expect or how to fully configure the query.

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 descriptions already cover 'directories' and 'prompt', and the description adds context that folders are used as query context. However, the 'model' parameter lacks any description in the schema, and the description does not clarify its role or accepted values, leaving a gap in the 67% coverage.

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 runs queries over project folders ingested as context, with the NotebookLM analogy reinforcing the concept. It is specific enough to distinguish from general chat or analysis tools, though it does not explicitly name alternatives or scope exclusions.

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?

Usage is implied through 'project folders ingested as context' and the NotebookLM analogy, suggesting it is for question-answering over codebases. However, it provides no explicit when-to-use guidance, no exclusions, and no reference to sibling tools like agy_chat or agy_analyze.

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 must fully disclose behavior, but it only mentions return style ('without conversational fluff'). It does not state whether the file is modified in place, permissions required, reversibility, or failure behavior, which is critical for a refactor tool.

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 short sentences front-load the action and result, with no filler. It is appropriately concise for the minimal info it conveys.

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 mutation-capable tool with three parameters, no output schema, and no annotations, the description is underspecified. It omits how changes are applied, what 'code changes' means (diff vs. full file), and how the optional model affects behavior. This falls well short of a complete tool description.

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 filePath and instructions, and the description aligns with those. The optional 'model' parameter is undocumented in both schema and description, and description adds no extra meaning beyond the schema's coverage of required 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 uses specific verb 'Refactor' and resource 'target file', clearly indicating the operation. It adds return behavior ('Returns code changes') and a tone guideline, but does not explicitly distinguish from sibling tools beyond the action verb.

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?

No guidance on when to use this tool versus alternatives like agy_analyze or agy_chat. It lacks context, prerequisites, or exclusions, leaving the agent to infer applicability from the name alone.

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 must carry the full behavioral disclosure burden. It mentions 'restricted sandbox environment' and 'safely,' but does not explain the nature of restrictions, execution behavior (e.g., synchronous or async), cleanup, permissions, or potential side effects. For a code execution tool, this is a significant gap.

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, front-loaded sentence that directly states the tool's purpose. It is concise and efficient, with no wasted words or redundant information.

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 sandbox execution tool with no annotations, no output schema, and low parameter coverage, a one-sentence description is highly inadequate. It lacks essential context about how to use the tool, what restrictions apply, and what to expect as a result, making it insufficient for safe and correct invocation.

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 only 33% (only 'prompt' has a description), and the tool description does not mention any parameters. 'cwd' and 'model' are completely undocumented, and the description adds no meaning beyond what the schema provides for 'prompt'.

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 action ('Run tasks') with a defined environment ('restricted sandbox environment') and purpose ('safely investigate code or execute shell scripts'). This distinguishes it from siblings like agy_task or agy_analyze by emphasizing the sandboxed, safe execution aspect.

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 clear context for when to use the tool: when you need to run code or shell scripts safely in a restricted environment. However, it does not explicitly mention alternatives or when not to use it, so it stops short of full exclusionary 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?

No annotations are provided, so the description carries the full burden of disclosing behavior. It states the tool scans for vulnerabilities but does not clarify whether files are modified, what the output looks like, or any permissions needed. The term 'audit' suggests read-only but is not explicit, leaving significant ambiguity.

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?

A single, direct sentence with no redundant wording. It effectively front-loads the core functionality and 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 lack of output schema and annotations, the description is incomplete for confident tool invocation. It omits return format, the purpose of the 'model' parameter, and any side effects, leaving critical gaps for the agent to navigate.

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 only 50% (filesToAudit is described, model is not). The description adds no new meaning to filesToAudit beyond the schema and completely ignores the 'model' parameter, leaving the agent without guidance on how to use it.

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 a specific verb ('Perform') and identifies the resource ('target files'), with a clear scope: scanning for OWASP vulnerabilities and credential leaks. This distinguishes it from sibling tools like agy_analyze or agy_refactor, making its purpose 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 implies use for security auditing but provides no explicit when-to-use guidance or alternatives. It does not mention when to choose this over agy_analyze or other analysis tools, leaving the agent to infer the context.

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 only lists capabilities (write code, analyze files, run commands) but omits critical behavioral traits such as potential file modifications, permission prompts, or the risks of the yolo flag. The description does not warn that executing commands could have 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.

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

The description is concise and front-loaded: two sentences that state the core purpose and key capabilities without unnecessary detail 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?

This is a complex task runner with six parameters and no output schema, but the description gives minimal context. It does not explain what the tool returns, how modes (accept-edits/plan) affect behavior, or how it relates to specialized siblings like agy_sandbox_agent or agy_autonomous_agent. The description is too sparse for a tool with this complexity.

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 provides descriptions for all 6 parameters, achieving 100% coverage. The description does not add further parameter-level detail beyond stating the tool runs tasks, so it stays at the baseline for schema coverage.

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: 'Run a task using Google Antigravity CLI (agy)' with a specific verb and resource. It also distinguishes the tool from siblings like agy_chat and agy_analyze by mentioning capabilities such as writing code, analyzing workspace files, and running commands.

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 implies usage for general task execution but does not explicitly state when to use this tool vs alternatives. It does not mention any exclusions or contrast with sibling tools, leaving the agent to infer the appropriate context.

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