vicp-registry

Tools

Token cost is the o200k count of the tools list JSON, including schemas, before any tool is called.

• search_cases

  Search VICP cases with structured filters. - query: free text matched against petitioner identifier, condition, package id, caption name and the AI case summary. - vaccine: canonical va

• get_case

  Get the complete curated record for one case by package_id. Includes the AI-assisted case summary, all extracted fields, attorney and special master, plus links: canonical page, plain-text so

• get_case_text

  Get the staged plain text of a case's public court documents. Long decisions can exceed 100 pages; the text is returned in chunks. Use offset_chars from the previous response's next_offset to

• get_filters

  List all valid filter values: canonical vaccine names (with case counts), condition categories, outcomes, special masters, and the dataset's date range. Call this before filtering so you use r

• get_statistics

  Aggregate statistics over the case corpus, optionally filtered. group_by: vaccine | condition | outcome | year | special_master Returns per-group: total cases, compensated count, compensation

• search_attorneys

  Search petitioner attorneys and firms by name. Returns per-attorney aggregate stats: case count, compensation rate, main vaccines, and firm. Useful for 'who has handled X cases' questions.

• get_attorney

  Get one attorney's profile: bio (when available), aggregate outcomes, and their most recent cases. Name is matched case-insensitively.

• get_special_master

  Decision patterns for one special master: outcome breakdown, top vaccines/conditions in their docket, and recent cases.

• search

  Search VICP vaccine-injury cases by free text. Returns result ids for use with fetch. (Alias of search_cases for deep-research clients; prefer search_cases when structured filters are availabl

• fetch

  Fetch the full record + source text of one case by its id (package_id). (Deep-research companion to search.)