Routine Run: write a routine

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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so safety is covered. The description adds genuine behavior beyond that: matching is against name or description, and each result includes its steps plus a link that installs it in the app.

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, both front-loaded: the search scope first, the return shape second. No filler or repetition of the title.

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 return-value burden and does so ('each comes with its steps and a link that adds it'), while the schema covers the empty-query case. Nothing needed to call this tool correctly is 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?

Schema coverage is 100% and the single parameter is already documented ('what it is for, in a few words; leave out to list all'). The description adds only that matching targets name or description, a modest gain over the schema baseline.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (searches) and a precisely scoped resource: the routines the app ships, not user-created ones. The examples (mornings, focus, on call, chores) make the resource concrete and separate it from the sibling how_to_write_a_routine.

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 scoping to app-shipped routines implies this is the lookup path when you want a ready-made routine rather than authoring one, but no when/when-not or explicit alternative (how_to_write_a_routine) is named. Usage is inferable, not 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?

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint=false, so the safety profile is covered. The description adds genuinely new behavioral information by describing what the call returns — the complete field set and action vocabulary of a routine — which is the key trait for a zero-argument reference 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, with the description of the output leading and the usage trigger following. No filler or repetition of the title, though the opening clause 'The JSON form Routine Run reads a routine in' is slightly convoluted.

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, no-output-schema reference tool, the description covers the two things an agent needs: what the payload contains and when to fetch it. It could be marginally stronger by naming an example field or step action, but nothing essential is 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 tool takes no parameters, so per the rubric the baseline is 4. There is nothing for the description to clarify about inputs, and it correctly spends no words on parameter semantics.

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 goes beyond the name by stating that the tool returns the JSON form Routine Run reads — every field and every action a step can take. That identifies it as a schema/reference tool rather than an action tool, which is enough to separate it from routine_link and find_starter_routine. It stops short of explicitly contrasting itself with those siblings, so it lands at 4 rather than 5.

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 gives an explicit trigger: 'Use it when writing a routine for routine_link, start_new_run or propose_new_routine.' That is concrete when-to-use guidance tied to named consumers of the output. It does not state when *not* to call it, which keeps it below a 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?

Annotations already declare readOnly/idempotent/non-destructive, so the bar is low; the description adds real context beyond them by explaining what the link does on an iPhone (opens the app's review screen, where the user chooses to save), that nothing is persisted here, and that the JSON should be passed along as a fallback.

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?

Purpose is front-loaded in the first sentence, and the remaining sentences each add operational value (link behavior, no persistence, hand off link plus JSON). The informal phrasing adds length but stays within a reasonable size.

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?

There is no output schema, so the description must describe the return, and it does: a link plus JSON with a note on the link's on-device behavior. Combined with annotations covering the safety profile, an agent has enough to invoke and route the result, though invalid-routine/error behavior is left unstated.

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% and the single parameter is already documented in the schema as accepting an object or JSON text. The description adds only an indirect pointer to how_to_write_a_routine for the format, so 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?

States a specific action (checks a routine, returns a link) and names the resource, and ties the input format to the sibling how_to_write_a_routine. It's distinguishable from find_starter_routine, but the phrasing is narrative rather than crisp, so it doesn't quite reach 5.

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?

Implicitly routes the agent to how_to_write_a_routine for the input form and tells it to hand the user both the link and the JSON. However, it never states when to call this tool versus find_starter_routine, nor any precondition beyond the referenced form, leaving usage largely inferred.

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