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 disclosure. It only states that the tool lists available agents, but discloses nothing about output format, ordering, completeness, authentication requirements, or potential error conditions. For a read-style listing tool this is a minimal but notable 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?

A single efficient sentence with no wasted words, front-loading the action and resource. It is well-structured for a simple tool, though it borders on under-specification rather than genuine 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?

Given zero parameters and no output schema, the tool is low-complexity and the description is arguably adequate for the agent to understand its basic action. However, the absence of any return-format details and the unexplained 'agy 账号' reference leave moderate ambiguity about what the response will look like.

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 baseline 4 applies. There is nothing for the description to clarify about parameters, and the schema already documents the empty input shape fully.

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 verb ('列出' / list) and a concrete resource ('可用的 agent' / available agents), making the tool's core purpose clear. It is distinguishable from siblings like antigravity_models and antigravity_jobs by the resource it targets, though it never explicitly contrasts itself with any sibling.

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 is given on when to use this tool versus alternatives, and no exclusions or context are provided. The phrase '当前 agy 账号' introduces an account scoping assumption but does not clarify how the agent is identified or when this tool should be preferred over related tools like antigravity_models.

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 behavioral disclosure. It reveals that it uses agy's Google account quota and continues the same session by default, but it omits potential side effects such as file modifications, command execution, or permission handling. For a tool that runs an AI agent, 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 sentence, very concise and front-loaded with the core action and quota note. It avoids verbosity, though given the tool's complexity it may be too brief. Still, it is efficient and well-structured.

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?

With 22 parameters and no output schema, the description is inadequate. It does not describe the return format, error behavior, or how session continuity works in practice. The quota and session notes are helpful but insufficient for an agent to use this tool correctly without diving into the schema.

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 all 22 parameters have descriptions. The tool description adds no extra parameter meaning beyond what the schema already provides; the mention of session continuation aligns with the session parameter but is already documented there. Baseline 3 is appropriate given the high 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 states a specific action: run a non-interactive question via the local Antigravity CLI and return the answer, and notes default session continuation and quota usage. However, it does not differentiate from sibling tools like antigravity_submit or antigravity_job, so an agent cannot tell them apart without inspecting their schemas.

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 is given on when to use this tool versus alternatives. It does not mention conditions, exclusions, or contexts where another sibling would be more appropriate. The only hint is the quota note, but that is about resource usage, not selection criteria.

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 present, so the description carries the full behavioral burden. It hints that 'forget' cleans up jobs, but it does not clarify whether this cancels a running job, only removes its record, or whether the action is reversible. This is a significant gap for a mutation-capable 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 a single, compact sentence that front-loads the verbs and names the resource. It is efficient and readable, though it could be slightly more structured by separating the operations from their requirements.

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?

With no annotations and no output schema, this description leaves real gaps: action-specific parameter requirements, the semantics and side effects of 'forget', and what 'get' returns are not covered. An agent can select the tool but may not invoke it correctly in all cases.

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's three verbs roughly map to the action enum values, and job_id is already documented in the schema as the id returned by antigravity_submit. However, the description does not clarify that job_id is effectively required for get/forget, nor does it explain the behavior of the default action.

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 specifies a concrete set of operations—查询/列出/清理 (get/list/forget)—against a named resource: background jobs started by antigravity_submit. This clearly distinguishes it from antigravity_submit itself, though it does not fully differentiate it from antigravity_status.

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 '由 antigravity_submit 启动的后台作业' implies this is the follow-up management tool for jobs created by antigravity_submit, giving some contextual clue. However, there is no explicit statement of when to prefer this over antigravity_status or when not to use it.

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 behavioral burden. The verb '列出' (list) implies a read-only operation and the account-scoping qualifier is useful context, but nothing is disclosed about the response shape, behavior when no models are available, or possible dependency on external calls.

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, front-loaded sentence conveys verb, resource, and scope with zero filler. Every word earns its place.

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, low-complexity list tool, the description covers the core call semantics, but with no output schema present it gives no indication of what the returned model list will look like. There is also no broader context hinting at how the result feeds into sibling tools such as antigravity_submit, leaving the agent to infer return structure.

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 is empty with zero parameters, so there is nothing for the description to elaborate. With 0 parameters the baseline is 4, and the description correctly stays silent on parameter details since none 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 uses a specific verb (列出/list) plus a specific resource (Antigravity models) and a scope qualifier (available to the current agy account), making the tool's function unambiguous. It does not explicitly distinguish itself from the sibling list-style tools antigravity_agents and antigravity_sessions, but the resource type is distinct enough that an agent can generally tell them apart.

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 call this tool versus the sibling tools. The description states only what the tool does, not when it should be used, and never names alternatives or conditions that would route an agent to antigravity_agents or antigravity_sessions instead.

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 behavioral disclosure. It proactively reveals key traits: the CLI answers directly, no turn is started, and no quota is deducted. This goes beyond the tool name and gives the agent confidence that calling it is read-only and cost-free.

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 with no redundant text. The purpose is front-loaded, and the behavioral caveats are compactly stated in one clause.

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 low complexity (0 parameters, no nested schema) and no output schema, the description sufficiently conveys what it does and its side-effect profile. An agent can safely invoke it based on this information; no crucial operational detail seems 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?

The input schema has zero properties and schema coverage is 100%, so the baseline for a 0-parameter tool is 4. The description adds useful semantic context about what dimensions are reported ('model group', 'weekly window', '5-hour window') even though no user-supplied parameters 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 uses a specific verb '查看' and identifies the exact resource—Antigravity remaining quota—with granularity details ('按模型组的周窗口与 5 小时窗口'). This clearly differentiates it from siblings like antigravity_ask or antigravity_submit, which concern model turns or submissions.

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 states that the CLI itself answers and that the tool does not start a turn or deduct quota, telling an agent this is the safe, no-side-effect way to inspect quota before choosing another action. It does not explicitly name alternatives or state when not to use it, so it stops just short of full 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 full responsibility. It states the actions (view/forget/list) but does not disclose side effects (e.g., whether forgetting is permanent, if it affects server state, or if it requires special permissions). The phrase 'list sessions that exist locally on the CLI' adds some behavior but is vague about output and consequences.

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 core actions. It is efficient and easy to parse, with no filler. Slightly unclear about the relationship between 'view or forget' and 'list local sessions', but acceptable.

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 (2 params, no output schema), but the description omits return values and edge cases (e.g., what happens when '*' is passed with forget). It provides enough to start using it, but not fully complete for an agent to know 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 description coverage is 100%, with both parameters already documented. The description adds no extra semantic detail beyond the schema, such as format examples or edge cases. It aligns with the schema's action/session definitions but doesn't enhance them.

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: viewing or forgetting server-tracked Antigravity sessions, plus listing CLI-local sessions. This distinguishes it from siblings which target jobs, status, etc. The resource (sessions) and verb (view/forget/list) are specific, though it doesn't explicitly contrast with sibling 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?

The description implies when to use it (for session management), but offers no explicit guidance on when not to use it or how it differs from alternatives like antigravity_job or antigravity_status. The context is clear but lacks exclusions or comparative routing.

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 carry the behavioral disclosure burden. It is transparent about what the tool reports, including a login probe and usage counters, but never states whether invocation is read-only, whether it can produce side effects, or what the login probe actually involves. That gap keeps it below a 4.

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 dense sentence packs the full report scope and a usage cue without filler. The long enumerative middle is slightly hard to scan, but every clause earns its place and the key troubleshooting guidance is placed at the end.

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 takes no inputs and the description lists the main report categories, so an agent can invoke it without risk and know broadly what to expect. Since there is no output schema, the description could be stronger on exact field names and formats, but for a zero-parameter diagnostic status tool this 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 100%-covered empty schema already fully documents the input surface. The 0-parameter baseline is 4, and the description needs no additional parameter detail.

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 reporting verb and enumerates exactly what is surfaced: AGY path, CLI version, working directory, agent visibility, daily calls/tokens, timing statistics, and login probe. This inventory plus the troubleshooting cue makes its role as an overall diagnostic status tool unambiguous and distinguishes it from siblings like antigravity_job, antigravity_quota, or antigravity_submit.

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 closing phrase '排查故障先看它' ('look at it first when troubleshooting') is explicit guidance on when to use the tool. It does not name specific sibling alternatives or state when not to use it, so it stops short of a full 5.

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 disclosure burden. It clearly discloses the core behavioral trait: the tool returns a job id immediately and runs in the background, and results come later via antigravity_job. It could add more (e.g., failure behavior, quota/session implications), but the essential async/non-blocking semantics are well covered without contradicting anything.

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 compact sentences with zero waste: purpose and return value first, then usage scenario, then follow-up tool. Every sentence earns its place and the most important information (background + job id) 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?

Despite 22 parameters and no output schema, the description covers the essential flow: what it returns (job id), when to use it, and how to retrieve results afterward. It doesn't describe failure modes or job lifecycle details, but the critical async workflow is complete enough for 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 description coverage is 100% — all 22 parameters have descriptive text in the schema itself, so the schema does the heavy lifting. The description adds a cross-reference (参数与 antigravity_ask 相同) rather than per-parameter detail, which is appropriate given the full schema coverage. Baseline 3 is correct.

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 (启动一轮 Antigravity 作业, start an Antigravity job) in the background and its immediate return value (job id). It clearly differentiates from siblings: it's the async counterpart to antigravity_ask and precedes antigravity_job. The purpose 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?

Explicitly states when to use it (长作业/深度分析/长文档 to avoid blocking) and the follow-up flow (之后用 antigravity_job 取结果), which routes the agent between the async submit tool and the result-fetching sibling. This is strong when-to-use guidance with a named alternative.

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