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

Annotations declare non-read-only, open-world, non-idempotent, non-destructive; the description adds the missing operational context an agent actually needs: credit consumption, the no-retry rule that follows from non-idempotency, provider refunds on failure, and the fact that submission returns immediately rather than blocking.

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 action and cost implication, and no sentence is pure filler. It is a dense run of facts, though, and the pricing clause plus the default-summary clause could be tightened without losing function.

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 still tells the agent what comes back (task id, statusTask, cost, metadata) and how to continue the workflow. For a 5-parameter async creation tool, nothing essential is left for the agent to discover by trial.

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 80%, so the schema already carries most parameter meaning; the description still adds value by naming the concrete defaults (seedance-2.0 / 10s / 9:16 / 720p) and pointing at the duration/model coupling. The 720p default is not represented in the schema, which is genuinely extra information.

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 (create ONE asynchronous video task) plus the key side effect (consumes account credits) and the immediate return payload (provider task id, statusTask, cost). It is unmistakably distinct from read-only siblings like get_video or list_models.

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 routing and sequencing: poll with get_video every 10–20 seconds, never auto-retry on network failure, and confirm pricing via list_models. Both when-to-use and when-not-to-do-something are spelled out.

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?

Goes far beyond the annotations: discloses directory creation, no-overwrite policy, 512 MB stream cap, 90-second network deadline, partial-file cleanup on failure, and constrained key forwarding across redirects. This is exactly the operational context an agent needs for a write-to-disk 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?

Densely front-loaded: purpose first, then return shape, then the write/overwrite/size/timeout constraints in short declarative clauses. No sentence is wasted.

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-param write tool with annotations and full schema coverage, the description covers return values (path, bytes, mime), limits, failure behavior, and auth handling. Nothing an agent needs to invoke it 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%, so both parameters are already documented with the same .mp4/absolute/not-exist semantics. The description reinforces but does not materially extend parameter meaning beyond what the schema provides, matching the baseline for high 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?

States a specific verb (Download) and resource (a succeeded task as an MP4) plus the destination (an absolute local output path). This makes it distinguishable from seedance_get_video, which would return info rather than write a file to disk.

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 'a succeeded task' establishes a clear precondition for use, and the disk-write behavior implies when this is appropriate. However, it never explicitly names an alternative (e.g., get_video) or states 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?

Beyond the read-only, idempotent, and non-destructive annotations, the description discloses rate-limit guidance, terminal-state semantics, signed URL expiry behavior, and retention limits. These are meaningful behavioral details an agent needs before polling.

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 and front-loaded with the core purpose, then appends polling and lifecycle guidance. The status enumeration is slightly run-on, but no sentence is wasteful.

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 read-only polling tool with annotations, a fully documented single parameter, and no output schema, the description covers statuses, terminal states, polling cadence, URL expiry, and retention. It gives the agent everything needed to call and interpret the tool correctly.

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?

With only one parameter and 100% schema description coverage, the schema already fully documents task_id as the ID returned by seedance_create_video. The description adds no parameter-specific meaning beyond 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?

The description uses a specific verb and resource: "Query task metadata and status" for a video generation task. It also enumerates the possible statuses, making the tool's role unambiguous relative to creation, listing, and download siblings.

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 clear operational guidance: poll no faster than once every 10–20 seconds, recognize terminal states, refresh expired signed URLs, and download within 24 hours. It does not explicitly name seedance_download_video as the alternative for downloading, so routing is implied rather than fully 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 cover readOnly/idempotent/non-destructive, but the description adds behavior annotations do not: no authentication performed, no credits consumed, and the key-presence-vs-validity caveat. Those are the exact traits an agent needs to interpret a successful health response correctly.

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 short clauses, front-loaded with the operation and return shape, then the caveat and the redirect to the sibling. Every sentence carries distinct information with no repetition of annotations or schema.

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 takes on the job of describing the return payload, which it does by listing the three returned signals. It also closes the main interpretive risk (treating a configured key as a working key), so nothing needed to call it 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?

Zero parameters, so there is nothing to disambiguate and the baseline of 4 applies. The description does not need to add parameter meaning and correctly does not pad the text.

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 ('Read-only connectivity check') and enumerates what the check returns: API health, dk_ key presence, and local config file path. It also distinguishes itself from seedance_list_models by naming that sibling as the authentication verifier.

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 when-to-use-this vs use-sibling rule: a configured key is not proof of validity, so use seedance_list_models to verify authentication. It also states the negative condition that this call does not authenticate or consume credits, removing ambiguity about its cost.

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 safety profile is covered. The description adds context annotations don't: authentication is required, disabled models surface as 404, and only currently enabled models appear. It stops short of describing pagination or rate limits, but adds real value.

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?

Four short sentences, front-loaded with purpose and followed by usage and return info; nothing is filler. Minor redundancy: 'Authenticated read-only request' restates annotation hints, but the authentication detail is arguably new information.

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-param, no-output-schema tool, the definition covers what it does, when to call it, the failure case, and roughly what comes back (provider data array with names and credit prices). A short note on array shape or fields would make it fully airtight, 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 zero parameters, so there is nothing for the description to disambiguate; baseline 4 applies. The description correctly does not invent 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?

States a specific verb and resource ('List currently enabled video models') plus what is returned (names and credit prices). This clearly separates it from siblings like seedance_get_video or seedance_create_video, which act on a single model/video rather than enumerating available models.

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?

'Call before generation' gives an explicit trigger condition, and 'disabled models return 404' explains a related failure mode. It doesn't name specific sibling alternatives, but the catalog-then-generate flow is clear enough to place the call correctly.

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?

Goes beyond annotations by disclosing the 5 MB cap, accepted formats, that it sends to the configured Seedance API (open-world), and the exact return payload (asset id, mime, byte count). Annotations already cover the read-only/idempotency profile, so the remaining gaps (auth specifics) are minor.

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 tight sentences: purpose/limit first, then return values, then integration instruction. Zero filler, and the actionable ordering guidance is front-loaded where it matters.

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?

No output schema exists, yet the description explains the return shape (asset id, mime, byte count), plus size/format constraints and how to use the ID. Nothing an agent needs to invoke or consume this tool 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 documents image_path as an absolute local path, so the baseline is 3. The description adds file-level constraints (allowed formats, 5 MB max) not encoded in the schema, which meaningfully improves the agent's ability to satisfy the parameter correctly.

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 (upload) and resource (local reference image) with concrete format (PNG/JPEG/WEBP) and size (5 MB) scope. An agent can immediately distinguish this asset-preparation step from siblings like create_video or download_video.

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 tells the agent how to consume the result: put IDs into reference_asset_ids in the same order as @Image1/@Image2 in the prompt. This is clear downstream guidance; it doesn't state explicit exclusions versus alternatives, but the workflow position is unambiguous.

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