Update to Socialinsider

{
  "description": "Audit how one social media profile performed over a period using Socialinsider data — followers and engagement trend, aggregate post metrics, post-type mix, and the posts that actually worked. Use when the user asks how a profile, account, brand or channel is doing, wants a performance review, monthly report, engagement-rate check, or asks why reach or engagement moved.",
  "included_files": [],
  "name": "profile-audit",
  "skill_md_contents": "---\nname: profile-audit\ndescription: Audit how one social media profile performed over a period using Socialinsider data — followers and engagement trend, aggregate post metrics, post-type mix, and the posts that actually worked. Use when the user asks how a profile, account, brand or channel is doing, wants a performance review, monthly report, engagement-rate check, or asks why reach or engagement moved.\n---\n\n# Profile audit\n\nProduce a performance audit for one tracked profile from Socialinsider data.\n\n## Rules\n\n1. **Call `socialinsider_get_date_context` first, always.** Your training data may be from an\n   earlier year. Never guess or hardcode a year. Use the returned `date_ranges` values verbatim\n   as the `date` argument (`{start, end, timezone}`, millisecond timestamps).\n2. **Never invent ids.** `id` and `profile_type` must come from `socialinsider_list_profiles`.\n   `projectname` is the project the profile lives in.\n3. **Aggregate, don't paginate.** Use `socialinsider_get_posts_summary` for totals and averages\n   and `socialinsider_get_top_posts` for rankings. Only fall back to `socialinsider_get_posts`\n   when the user needs individual post records.\n4. **Read-only.** This skill never calls a write tool (`create_*`, `add_*`, `remove_*`).\n\n## Steps\n\n1. `socialinsider_get_date_context` — get today's date and ready-made ranges.\n2. `socialinsider_list_profiles` — resolve the profile the user named to its `id`, `profile_type`\n   and `projectname`. If several profiles match, ask which one before continuing.\n3. Pick the window. Default to `last_30_days` when the user didn't say. If the user asks about a\n   trend or change, also fetch the immediately preceding window of equal length so you can compare.\n4. Gather, in parallel:\n   - `socialinsider_get_profile_data` — daily followers, engagement rate, reach, impressions.\n   - `socialinsider_get_posts_summary` — true post count, totals, averages, breakdown by post type.\n   - `socialinsider_get_top_posts` — `metric: \"engagement\"`, `limit: 5`.\n5. If the profile is `instagram_profile_own`, add `socialinsider_get_instagram_stories` for the\n   same window. Skip it silently for any other profile type — it is unavailable there.\n\n## Output\n\nA short written audit, in this order:\n\n- **Headline** — one sentence: direction and size of the move (\"engagement rate down 18% MoM on\n  flat follower growth\").\n- **Numbers** — a compact table: followers (start, end, delta), posts published, total engagement,\n  average engagement per post, engagement rate. Add the prior-period column when you fetched one.\n- **What worked** — the top posts, with the metric that made them top and what they have in common\n  (format, topic, length, posting time). Name the pattern, not just the posts.\n- **Post-type mix** — which formats the profile leans on versus which formats earn engagement.\n  Call out a mismatch when it exists.\n- **Recommendations** — at most three, each tied to a number you just showed.\n\nState the date range and timezone you used. Flag any metric the API returned as empty rather than\nreporting it as zero — for non-owned profiles, reach and impressions are often simply unavailable.\n"
}