Hello world, this is a test.
'","exampleUseCases":[{"title":"Mid-task context window shrink","prompt":"I have a big HTML blob from a scraped page that I need to fit into my context window — can you convert it to markdown and tell me the approximate token count so I know if it'll fit?"},{"title":"LLM-ready content extraction","prompt":"Take this HTML article fragment and convert it to clean markdown so I can feed it directly into my summarization step without all the tag noise."},{"title":"Token budget estimation before processing","prompt":"Before I send this HTML to the model, convert it to markdown and give me an approximate token count so I can decide whether to chunk it or send it whole."}],"resultDescription":"A JSON object with three fields: 'valid' (boolean indicating whether conversion succeeded), 'markdown' (the converted Markdown string), and 'approx_tokens' (an integer approximation of the token count of the resulting markdown).","failureModes":["Malformed or non-HTML input may still be processed best-effort but 'valid' may be false","Very large HTML fragments may produce inaccurate token counts","Dynamic or JavaScript-rendered content won't be executed — only static HTML is parsed","Deeply nested or exotic HTML structures may produce imperfect markdown","Empty input string returns minimal or empty markdown"],"whenToPreferThis":"Choose this endpoint when an agent needs a lightweight, pay-per-call HTML-to-Markdown conversion with an approximate token count included — especially useful mid-task for context window management. It is not a browser or renderer, so prefer it for static HTML fragments rather than JavaScript-heavy pages. Pair with text-metrics endpoints if precise token counts are required.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-15T10:25:27.029Z","isFirstParty":false}