taste skill

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. "Read" implies a safe, non-mutating operation, but the description says nothing about error behavior (missing slug/path), output format, or any size/path restrictions, leaving meaningful gaps for an unannotated 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 tight sentences: the operation and its examples come first, followed by the triggering condition. Nothing is redundant or padded.

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 two-parameter read tool with full schema coverage, the definition is nearly sufficient; the only missing piece is what the call returns (file content) and how failures surface, which is a minor gap given the trivial read semantics.

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 both parameters are already documented in the schema. The description's path examples mirror the schema's own examples and add no new syntax, format, or constraint information, making the baseline 3 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?

States a specific verb (read) and resource (a supporting file inside a skill) and reinforces it with two concrete path examples. It clearly separates itself from the skill_* siblings, which appear to load whole skills rather than files within them, though it doesn't explicitly name an 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?

"Use after loading a skill that referenced an attachment" gives explicit timing and a triggering condition that tells the agent when this tool applies. It lacks an explicit when-not clause, but for a narrow read tool the context 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, the description carries the full burden and does disclose real process behavior: it draws from a curated library, enforces a house style via a 'real-browser inspector', and runs a screenshot look-and-fix loop. What is missing is whether invoking it merely returns instructions or actually mutates project files, and whether permissions or destructive changes are involved.

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 purpose is front-loaded, which is good, but the body is one sprawling, keyword-stuffed sentence ('in any stack (HTML/CSS, React, Next.js, Vue, Svelte, Astro, Tailwind)') and leans on promotional language ('anti-slop', 'award-level', the 247-gallery statistic) that crowds out the operational facts. Some trigger list earns its place for routing, but the whole is bloated.

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, no-output-schema tool, the description is largely sufficient to decide when to invoke it. It is incomplete on the one thing that matters most here: what invoking actually yields (a prompt/instruction payload vs. an executed workflow) and how it relates to the sibling read_skill_file.

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 zero parameters and schema description coverage is 100%, so there is nothing for the description to disambiguate. Baseline 4 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 clearly states what the tool is: a frontend art-direction skill that pulls researched design directions from a library and enforces a no-slop house style. That is a specific verb+resource, far better than a tautology. It loses the top mark because it never distinguishes itself from the crowded sibling set (skill_design_taste_frontend_v1, _v2_experimental, skill_gpt_taste, skill_stitch_design_taste), so an agent cannot tell which taste skill to pick.

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 context is unusually well spelled out — it lists concrete triggers (build/design/redesign/restyle/polish/review a landing page, portfolio, hero; user complaints like 'looks generic' or 'AI-made') and covers both marketing pages and app screens. There are no exclusions or alternatives named, however, so the agent gets no guidance on when a sibling skill would be the better choice.

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