Ingest any arbitrary URL on the fly, clean the HTML/document, run Workers AI structured extraction, and return verified JSON.
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | Target webpage or document URL | |
| instructionPrompt | Yes | Instructions on what exact data to extract into JSON |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full disclosure burden and does reveal a genuine pipeline: remote fetch, HTML/document cleaning, Workers AI extraction, and JSON verification — informative beyond the tool name. However, it omits edge behaviors such as failure on inaccessible URLs, document size/timeout limits, or what 'verified' actually entails.
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?
A single 21-word sentence that front-loads the core action and follows logical pipeline order (ingest → clean → extract → verify). Every clause earns its place; there is zero redundancy or 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?
For a low-complexity tool (2 required params, no nested objects), the description covers input, process, and output in one compact statement. But with no output schema and no annotations, it leaves the exact JSON structure undefined and says nothing about failure modes or input constraints for a tool that accepts 'any arbitrary URL'.
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% — both url and instructionPrompt already have meaningful descriptions in the schema. The description's pipeline language ('structured extraction', 'verified JSON') loosely maps to instructionPrompt's role but adds no parameter-specific detail beyond what the schema already provides, so the baseline of 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?
The description states a specific verb-resource pair ('Ingest any arbitrary URL') and a clear multi-step pipeline: clean HTML, run Workers AI structured extraction, return verified JSON. 'Any arbitrary URL' implicitly distinguishes this as the general-purpose tool against the domain-specific siblings (b2b pricing, clinical trials, zoning, etc.), though it never names that contrast explicitly.
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 'any arbitrary URL' implies this is the catch-all tool for URLs not covered by the specialized refinery siblings (pricing matrices, SEC 10-Ks, zoning, etc.). However, there is no explicit when-to-use vs. when-not-to-use statement, no named alternatives, and no exclusion criteria for documents the tool cannot handle.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Several tools form near-overlapping pairs: b2b_pricing_matrix vs custom_b2b_saas_pricing_matrix, dev_breaking_changes vs custom_dev_sdk_breaking_changes, and municipal/real-estate zoning vs regulatory_compliance. While individual descriptions differ, an agent would often have to guess which variant applies.
Names share a refinery_ prefix and use snake_case, making them mostly predictable and readable. The custom_ qualifier is used inconsistently—custom schema vs custom URL—and refinery_refine_custom_url/semantic_search break the otherwise noun-object pattern, but these are minor deviations.
13 tools is reasonable for a data-refinery platform covering multiple vertical schemas. The redundancy between base and custom_ variants and overlapping compliance tools makes it slightly heavier than necessary, but not excessive.
The surface covers refining arbitrary URLs, semantic search, and many domain-specific extraction schemas. Missing are database/schema management, document-level retrieval/update/delete, and clear parity between base and custom variants, so agents may hit dead ends when managing or verifying refined data.