Agent skills we publish
A skill is a file your coding agent loads into its own context when a task matches. These are ours, written from production incidents rather than from documentation, and they are free under MIT. 4 of them so far.
Supabase RLS Audit
Whether row level security is protecting anything, or only looking like it does.
Agent skillPostgREST Silent Failures
Requests that succeed and do nothing. The client side half of the same problem.
Agent skillNetlify Deploy Traps
The deploy said Published. That is not the same as the right files, or the right rules, going out.
Agent skillAI Front Desk
An automated front desk does not fail by throwing. It fails by saying something confident and wrong to a real customer.
Install all of them
Installs into the user level skills directory, which is the one Claude Code actually reads. Cursor reads it too, so one copy is enough for both.
npx skills add halltony85-source/clickflame-agent-skills -g -y
Read it before you install it
The complete text of every skill is printed on its page here, because a skill is instructions injected straight into your agent's context. The file is the payload. Read it before you install it, here or anywhere else.
Free, and staying that way
All work is by donation only. Nothing here is behind a payment, a trial or an email capture. If something we published saved you an afternoon, that is what the tag is for.
Donate on Cash App $clickflameLast reviewed 2026-09-11.