Get docket entries for a bankruptcy case. Use this when you have a case number. Steps: 1. Call with the short case number (e.g. 26-00712, 22-12345). 2. Different courts can use the same case number. If the number exists in more than one court, provide court_state (2-letter state abbreviation, e.g. PR for Puerto Rico, TX for Texas) or court_id. If you get no results or the wrong case, ask the user for the state and pass court_state. 3. The tool resolves the case number (and optional court_state or court_id) to the case, then returns docket entries plus dates so you can say how recent the information is. Returns: JSON with entries (array of itemNumber, itemText, itemDate, …), latestEntryDate, dateDocketUpdated. Response includes: - entries: list of docket items (itemNumber, itemText, itemDate, etc.). - latestEntryDate: date of the most recent docket entry in our data. - dateDocketUpdated: from the case table; when this case's docket was last synced to the database. Use this to answer "how recent is this docket information?" live_update: if True (subscribers: no extra charge), run a one-time refresh from source before returning (slower; may not be enabled). This does not create ongoing monitoring.

Returns aggregate platform statistics. Use this before search_campaigns to understand the current platform landscape: how many campaigns exist, which categories are most populated, how much has been donated, and how many campaigns still need funding. No parameters required.

Search for application guidance: deadlines, forms, how to apply, and scheme rules.

Ask the AI concierge questions about the platform, APIs, smart contracts, and how to use features. No api_key required.

Estimate how much revenue a seller could earn from their API on the platform.

[$0.003 USDC per call (x402)] Platform accuracy scores and calibration analysis. How often do markets priced at 70% actually resolve Yes 70% of the time? Identifies which platforms are most reliable, which categories are well-calibrated, and where systematic biases exist. Powered by PROWL historical analysis. Use this to answer 'how accurate are prediction markets?' or 'which platform is most reliable?'