Human For AI

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

Annotations already declare readOnlyHint=true and openWorldHint=true, setting the safety baseline. The description adds valuable behavioral context beyond annotations: the status lifecycle, seen_by_operator_at semantics, ETA, operator notes, and the JWS receipt verification process. No contradiction with annotations exists.

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 front-loaded with the main purpose ('Look up a submitted task by its task_id') and then efficiently lists the response fields in three sentences. While information-dense, it contains no filler or redundant phrases, and the structure is logical. It could arguably be two sentences, but the detail justifies its length.

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 output schema, the description carries the full burden of explaining return values. It thoroughly covers status, status history, seen_by_operator_at, eta, operator notes, receipt, and deliverable_sha256, including a verification URL. For a single-parameter lookup tool, this is comprehensive and complete.

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 for the only parameter (task_id), including a description and example. The tool description does not add any extra parameter explanation, so it neither enhances nor detracts from the schema. Baseline 3 is appropriate given the schema's completeness.

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: 'Look up a submitted task by its task_id.' It specifies the resource (submitted task) and the action (look up), and the detailed return fields further define its scope. This distinguishes it from sibling tools like submit_human_task (submission) and message_human_operator (messaging).

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 after task submission (e.g., 'Look up a submitted task') and details what information is returned, providing clear context. However, it does not explicitly state when to use this tool over alternatives or mention exclusions, so it stops 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?

Annotations already declare readOnlyHint=true and openWorldHint=true, and the description complements these by detailing the manifest contents and elaborating on the open-world semantics ('catalog is examples, not limits'). It adds useful context beyond annotations without contradicting them.

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: one sentence summarizes the manifest contents, and a second provides usage guidance and the open-world caveat. Every sentence earns its place, and the most critical instruction ('Call this first') appears prominently.

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 no-parameter read-only discovery tool, the description fully covers what the tool returns, when to use it, and how to interpret the results. Combined with strong annotations and no output schema, no additional context is needed.

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 schema coverage is 100%, so there is no parameter burden for the description to carry. The baseline of 4 applies because no parameter documentation is needed.

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?

Uses a specific verb ('Fetch') plus a clear resource ('Human For AI manifest') and enumerates the exact contents (services, operator profile, response times, task types, policy). This clearly distinguishes it from sibling tools like submit_human_task or message_human_operator, which are action-oriented.

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?

Gives explicit guidance to 'Call this first' before deciding to hire the human, establishing a clear ordering. It also notes unlisted needs are welcome, but it does not explicitly name alternative tools or state when not to use this tool, so it falls just 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?

Annotations already signal a side-effectful, open-world, non-idempotent operation. The description adds meaningful behavioral context beyond annotations: reply_to is REQUIRED and is 'the only way the operator can answer,' implying asynchronous human interaction and a dependency on a reachable email. No contradiction with annotations.

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 tight sentences with no wasted words. The core action is front-loaded, and the critical reply_to constraint is stated immediately after the purpose. Every clause 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 simple messaging tool with full schema coverage, annotations, and clear sibling context, this description is complete. It covers purpose, usage guidance, and the key operational constraint (reply_to). No output schema is needed, and no nested objects are present.

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. The description adds extra semantic value by explaining that reply_to is the only channel for the operator to reply, reinforcing the schema's 'REQUIRED' and 'MX-checked' notes. It does not elaborate on from or subject, but the schema already describes those 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 opens with a specific verb and resource ('Send a free-form message to the human operator') and lists concrete use cases ('questions, scoping, custom or recurring projects'). It also distinguishes itself from submit_human_task by explicitly stating this is for 'anything that is not yet a ready-made task.'

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 clear context for when to use the tool ('anything that is not yet a ready-made task') and highlights the critical reply_to requirement. It does not explicitly name alternative sibling tools like submit_human_task or check_task_status, but the 'not yet a ready-made task' phrasing implies the alternative for ready-made 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?

Beyond annotations (readOnlyHint=false, etc.), the description discloses key behaviors: human review introduces delay, push-notification on submission, seen_by_operator_at field, MX-check requirement, and the status_poll one-per-day limit. This adds substantial non-obvious context.

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 yet each sentence delivers actionable information: purpose, non-instant execution, status tracking, cost, email validation, and the no-mailbox alternative. It is well-organized and front-loaded with the core 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 9-parameter tool with no output schema, the description covers all essential behavioral and delivery aspects. The schema handles parameter details, while the description provides the review workflow, delivery options, and constraints needed for correct usage.

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 rich per-parameter descriptions. The tool description repeats some details (contact_email MX-check, status_poll budget) already in the schema, so it adds marginal value beyond the structured definitions.

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 opens with 'Submit a task for the human operator to perform in the real world,' clearly identifying the verb+resource and distinguishing it from siblings like check_task_status or message_human_operator. It also clarifies immediate task_id return vs. human review, adding specificity beyond the title.

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 explicit when-to-use and alternative guidance: 'No mailbox? Set delivery to status_poll instead' and references check_task_status for monitoring. It also notes the pilot is free and explains the status_poll budget, giving clear contextual boundaries.

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