Stringer CleanExtract

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

Annotations only carry openWorldHint, so the description carries the behavioral load and does so richly: billing model (charged only on usable content, first 3 free then USD 0.05), enumerated failure codes that are not charged, the fallback_used signal, and the success quality fields. These are non-obvious traits an agent needs to plan cost and retries.

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 differentiating comparison, and every sentence carries weight (benchmark, sequencing, billing, error taxonomy). The error-code enumeration is dense but is genuinely decision-relevant, so only a small deduction for 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?

No output schema exists, so the description must describe returns — it does, naming extraction_quality, extraction_content_ratio_band and fallback_used. Combined with failure modes and billing, an agent has everything needed to call and interpret this tool.

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 three parameters are already documented in the schema; the baseline is 3. The description adds only the origin of section ids (from clean_extract_outline) and document-order return, which is useful but marginal over the schema.

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 (extract), resource (full page or selected sections from URL/raw HTML), and output form (clean Markdown with navigation, scripts and chrome removed). Also distinguishes itself from a raw fetch, so an agent can tell it apart from generic retrieval.

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 says to use it instead of a raw fetch for documentation, articles, RFCs or reference pages, and gives the sequencing rule: call clean_extract_outline first on a long page, then pass only the needed section ids. This is clear when-to-use and how-to-compose guidance referencing the sibling tool.

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 well beyond the readOnly/openWorld annotations by disclosing the billing model (never consumes the free extraction allowance; clean_extract charged only on usable content), the enumerated uncharged failure reasons, and the fallback_used signal. This is exactly the behavioral context an agent needs to sequence calls safely.

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 free-outline framing and output fields, then the routing advice, then billing and error semantics. Dense but every sentence carries information; the error-code list is the only slightly bulky element, yet it is genuinely useful.

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 and a single trivial parameter, the description compensates by naming the returned fields, the failure modes, and the fallback signal. An agent has everything needed to call and interpret this tool.

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?

Only one parameter, and schema coverage is 100%, so the schema already documents url_or_html fully. The description adds no syntax or format detail beyond the schema (the schema itself even says convert to Markdown, which is not quite the tool's job). 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 (outline) and resource (web page sections) and enumerates what the output contains: section ids, headings, byte sizes, and a fingerprint, explicitly excluding body text. This is distinguishable from the clean_extract sibling 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?

Explicitly routes usage: 'Use it first to see what a long page contains and which sections are worth reading,' and names the alternative flow ('A later clean_extract call can select sections'). The when-to-use and the follow-on tool are both stated.

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