Kodecz

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

Annotations already declare readOnlyHint and closed-world, and the description adds meaningful context beyond them: it is non-rendering (cheap), it reports per-problem frames, and `ok: false` is the failure contract. It stops short of saying whether the 'pictures of the worst frames' are written to disk, which matters for a tool marked read-only.

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?

Front-loaded with the core action and the 'without rendering' qualifier, then the rule list, then the return/failure contract. The rule enumeration is long but each item is a distinct check, so it earns its space; only the run-on list construction costs a point.

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 correctly takes on the return-value burden by naming the JSON result, per-problem frames, and worst-frame images. It is nearly complete for a two-param read-only check, with only the on-disk side effects of the images 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 both parameters carry their own descriptions (folder with index.html, id from the JSON script block), so the schema does the work. The description adds no additional parameter semantics, 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 and resource ('run automatic checks on every frame') and enumerates the exact rule set being validated. The phrase 'without rendering' cleanly separates it from the sibling `render`, so an agent can pick it without opening a schema.

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?

'fix and check again' implies an iterate-until-ok workflow, which is real guidance. However, it never says when to prefer this over `preview` or when to skip it and go straight to `render`, leaving the sibling routing to inference.

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 cover the safety profile (readOnlyHint=true, openWorldHint=false). The description adds genuinely useful behavior the annotations cannot convey: lossless PNGs, pixel-identical to the render, limited to one to two seconds of footage, and returned as images rather than paths.

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, front-loaded with the core action and return format. Every clause earns its place, including the closing rationale for using the tool.

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 all three required parameters fully described in the schema and no output schema, the description fills the main remaining gap by explaining that images are returned directly. Minor omissions: no note on cost, latency, or behavior with out-of-range frame numbers.

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 all three parameters documented including frame-number examples, so the schema carries the load. The description adds nothing about parameters, though it does note the 1-2 second scope, which loosely constrains the frame count. Baseline 3 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?

States a specific verb and resource ("Draw some frames of a Kodecz video as lossless PNGs") and the return type ("return them as images"). It implies the relationship to `render` by saying the pixels match the render, but never names `render` or `check` to explicitly draw the boundary.

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 it to look at what you wrote instead of guessing" gives a concrete motivation and context for choosing this tool over blind edits. It stops short of stating when to prefer `render` or `check` 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?

Annotations already declare it is not read-only, is idempotent and non-destructive, and the description usefully adds the GPU/NVENC-vs-CPU/x264 selection rule and the return shape ('JSON result line'). However, it never says what happens to an existing `output` file, whether the render is long-running, or what the result line contains — notable gaps for a file-writing mutation.

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 action and encoding rule come first, the return value and prerequisite follow. Zero filler, and the prerequisite is placed where it will be read before invocation.

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 8 parameters, a mutation profile, and no output schema, the description supplies the key missing pieces: encoding fallback behavior, the return format, and the `check`-first requirement. Overwrite semantics and expected runtime remain unstated, which keeps it from being 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?

Schema description coverage is 100%, so all 8 parameters are already documented in the schema, including the GPU-only constraint on `hevc` and `alpha`. The description's GPU-vs-CPU sentence largely restates the `cpu` parameter's own text, adding little beyond what the structured field already provides.

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 (render) and resource (a Kodecz video folder) plus the output format (MP4) and encoder selection logic, so the agent knows exactly what happens. It does not distinguish itself from the `preview` sibling, so it falls just short of a 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?

Gives an explicit prerequisite sequence ('Run `check` first'), which is actionable context an agent would otherwise miss. It offers no guidance about when to choose `preview` instead, so it is clear but not fully discriminating.

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