Update to Conveo

{
  "description": "Use when someone asks about their Conveo studies, wants to design a new interview study, check recruitment, or analyze what participants said in Conveo interviews.",
  "included_files": [],
  "name": "get-started",
  "skill_md_contents": "---\nname: get-started\ndescription: Use when someone asks about their Conveo studies, wants to design a new interview study, check recruitment, or analyze what participants said in Conveo interviews.\n---\n\n# Working with Conveo\n\nConveo runs AI-moderated video interviews. A **study** has a topic guide made of\n**sections**, each holding **questions**. Participants are recruited into the\nstudy, and every completed conversation is an **interview**.\n\n## Find the study first\n\nMost requests name a study loosely (\"my coffee study\"). `list_studies` has no\nname search: it pages through studies newest first, so read the titles and\nmatch them yourself. If more than one study matches, ask which one — never\nguess between them.\n\n## Analyzing results\n\n- `query_study_data` takes no study ID; it always runs inside an analysis\n  session. Create one with `create_analysis_session` (pass the study's ID in\n  `studyIds`), then pass the returned session ID and the question to\n  `query_study_data`.\n- Ask follow-up questions in the same session — it keeps the conversation.\n  Each query runs a full analysis, can take a few minutes, and counts as one\n  analysis query for the organization.\n- The answer is markdown `text`, plus `quotes`, `tables` and `charts` when the\n  analyst produced them. Replace a `[ref:message:<id>]` marker with the quote\n  whose `messageId` matches, and `[chart:<id>]` / `[table:<id>]` with the item\n  whose `id` matches; drop a marker with no match. Never show raw markers.\n- No tool counts how many participants support a single finding. Don't invent\n  one — rely on what the analyst's text says.\n- To resume earlier work, find the session with `list_analysis_sessions` and\n  re-read it with `get_analysis_session_messages` instead of asking again.\n- If the study belongs to an Always-On storyline (a wave-based program), find\n  it with `list_storylines` and read its findings with\n  `list_storyline_insights` before running a new analysis. Ordinary studies\n  have no storyline.\n\n## Finding clips\n\n`search_clips` matches a literal keyword (case-insensitive) in the quotes\nConveo has already extracted from completed video interviews — not full\ntranscripts, and not by meaning — so try related words too (price, cost,\nexpensive). No results doesn't mean nobody said it: fall back to\n`query_study_data`. Results carry the quoted text and interview reference, but\nno link to the moment in Conveo — don't promise one.\n\n## Designing a study\n\n- Create it with `create_study`. Only `title` is required. It returns the\n  study's ID and slug, not a link.\n- Every new study is section-based. Build the interview with `add_section`,\n  then add questions with `add_question`, passing the section's ID as\n  `sectionId`. `type` defaults to open-ended.\n- A new study comes with a \"Screener\" section (`type: \"screener\"`) and one\n  empty, untitled interview section (`type: \"userCreated\"`). Find the untitled\n  one with `get_study` and name it with `update_section` rather than leaving\n  it blank. Don't rename the Screener or put interview questions in it.\n- Research objectives are analytical only: they record what the analysis\n  should answer and hold no questions. Don't use `add_research_objective` to\n  structure the interview, and never pass an objective ID to `add_question`\n  in a section-based study — the call fails. (Some older studies have\n  `isSectionBased: false`; there `get_study` lists the topic guide under\n  `researchObjectives`, and questions are added with that objective's ID.)\n- Write questions that are open and non-leading (\"Tell me about the last time\n  you…\" rather than \"Did you like…\").\n- These tools cannot launch a study or recruit anyone. Tell the user to open\n  the study in Conveo to review and launch it.\n\n## Recruitment\n\n`get_recruitment_overview` reports interview counts by status (completed,\nongoing, screened out, abandoned) and each quota group's configured targets.\nIt does not report how full each quota is, so describe overall progress and\nthe quota setup — never per-quota progress.\n\n## What Conveo does not share\n\nConveo never returns participants' contact details. If asked for emails or\nphone numbers, explain that and offer what is available instead.\n"
}