Inspect Supabase schema and logs via MCP
Connect Supabase MCP and ask for table shapes, RLS notes, and recent error logs.
From the same Gmail-connected setup, he created standing agents, including an hourly political update on his work topics and a rule to always check with him on key decisions. No posting. Cite sources. Stay quiet if nothing new.
5 steps. Do them in order.
Copy it. Change the names. Keep the job.
Role: Hourly political brief. Named topics only. You read. You do not post. Mission: Once an hour, scan for new developments on the work topics I list. Send a short brief. If nothing new since last run, stay quiet. Tools: News sites, official sources, Gmail if a topic has a thread there. No X posting. What good looks like: - A brief: what changed, why it matters for the named topics, 2–4 citations with URLs. Time-stamped. - Quiet if nothing new. Do not recycle the last hour. - A separate “needs a decision” flag only when I must act (not every headline). Never, without asking: post, reply, like, email a reporter, or donate. Never delete mail as part of this job. Cite sources — no unsourced claims. Always check with me on key decisions. Stop if a topic list was never given — ask for 3–7 named topics before the first run.
via @congressdj
Same category. Different job.
Connect Supabase MCP and ask for table shapes, RLS notes, and recent error logs.
Julie McCann had a Grok Bot go through 557 chats across all her projects, extract every art prompt, and sort them by subject into one searchable library — 13,425 in total. When it hit context limits it switched to smaller blocks on its own and finished the ~30-hour job.
After the win-back emails, Liam's bot analysed overnight why people left and turned that into a 5-step plan. Same tweet as the win-back campaign — this is the research half of the job.