Web Page to Markdown (Tanod)

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

Goes well beyond the annotations (readOnly, idempotent, openWorld, non-destructive) by disclosing pricing (USD 0.002/URL, no free tier), that over-limit and all-failed calls are not charged, timing (1-5 s, 20 s bound), per-URL error status semantics, and an untrusted-data warning. This is exactly the operational context annotations cannot carry.

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 purpose and alternative, then packs pricing, limits, timing, and safety into a dense single block. Slight redundancy ('Static fetch only, no JavaScript' overlapping the Markdown/no-JS statements) prevents a 5, but no sentence is filler.

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?

Complete for an open-world read tool: it covers scope, limits, pricing, latency, error semantics, and prompt-injection caution, and the output schema handles the return shape. Nothing an agent needs before calling 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 each field (including the pass-through no-op options) is already documented in the schema as 'accepted and ignored.' The description's summary of which parameters matter (urls, text) and that highlights/summary are ignored largely restates the schema, so the 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 ('Get the contents of 1-10 URLs: web page to text') plus the return format and the sibling it is not (render_page for JavaScript). An agent can distinguish it from render_page and extract_pdf/html_to_text 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: 'For one page with JavaScript use render_page' and 'Static fetch only, no JavaScript.' It also gives the constraint that more than 10 URLs is a 422, so the agent knows the operating envelope and the alternative for dynamic pages.

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 cover the safety profile (readOnly, idempotent, non-destructive, closed-world), and the description adds substantial behavior beyond them: content-stripping rules (scripts, styles, nav, footers, forms, embeds dropped; main/article preferred), isolation in a resource-limited child process, 422 html_too_complex thresholds, latency range, and that failed-limit calls are not charged. The untrusted-data warning about page text and on-chain strings is a genuinely valuable operational caveat.

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 purpose then routing, with limits, pricing, and caveats in descending order of importance. It is dense and a few clauses duplicate the schema ('nothing is fetched' appears twice, the title sentence is restated), which keeps it short of a 5 for a tool of this simplicity.

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?

Given an output schema exists, return values need not be explained; the description instead covers everything else an agent needs: input constraints, error codes, cost, free-tier quota, execution environment, and latency expectations. Nothing material for correct invocation 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 description coverage is 100%, including the key html clarification (document itself, max 2 MB, nothing fetched) and base_url's relative-link behavior, so the description largely repeats structured data. It reinforces rather than extends the schema, which lands at the baseline 3.

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 ('Convert HTML to Markdown or plain text') and immediately names the two sibling tools it must be distinguished from: render_page for fetching a URL and markdown_to_html for the inverse operation. An agent can pick the right tool without opening any 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?

Explicit when-to-use routing: 'To fetch a URL use render_page; for the reverse use markdown_to_html', plus the negative constraint that the input is HTML, never a URL, and nothing is fetched. This is exactly the disambiguation a large sibling list (render_page, extract_pdf, get_page_meta, convert_document_to_markdown) demands.

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