Grow your AI search visibility (GEO)

Every week, do the following:
1. Call get_workspace for the company profile, and list_keywords to map
   "{BRAND_NAME}" and the competitors: {COMPETITORS}.
2. Find the surfaces AI engines actually cite. Call list_mentions filtered
   to reddit, hackernews, youtube, and news — last 30 days, competitor and
   category keywords, sorted by relevance. You're looking for:
   - "best X tools" listicles and roundups
   - recommendation questions ("what do you use for...")
   - comparison threads naming competitors
3. For each surface found, check whether {BRAND_NAME} appears in the body.
   Split into: mentioned (with sentiment) vs missing.
4. Using the analytics tool, compute brand vs competitor mention volume on
   those citation-heavy platforms over the last 90 days — this is your
   leading indicator for AI visibility. Note the week-over-week trend.
5. Prioritize the gaps: threads and lists where competitors are present
   and {BRAND_NAME} is missing, ranked by likely citation value (platform
   authority, engagement, recency). Take the top 5.
6. For each of the top 5, produce:
   **Surface:** {url} ({source}, engagement level)
   **Who's there:** which competitors are named
   **Action:** reply in thread / request inclusion in the list / publish a
   comparison page that answers this query
   **Draft:** the actual contribution — genuinely helpful, answers the
   question asked, discloses affiliation with {BRAND_NAME}
7. Post the digest in chat: the volume trend from step 4, the top-5 gap
   actions with drafts, and what to remeasure next week. If you track AI
   visibility with a dedicated tool, note this week's numbers next to the
   mention trend.
Do not post anything anywhere. Drafts are for a human to review and send.