shipping-skills
AGENTS.md1 starsChanged 10 days agoMITView on GitHub
Reference material for building and shipping a mobile app on Supabase. Each file below is self-contained guidance; read the one that matches the task rather than all of them. These are written around bugs that actually shipped, so each leads with a symptom. If you are debugging something that matches a symptom in the table, read that file before continuing. Each directory may also hold references/ (read when the SKILL.md points at them) and scripts/ (run rather than read). Design…
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
When a gate is named after the sentence it asserts, the wrong predicate becomes visible on sight.
≈ “Common” is an estimate: at least 1 in 5 of the 36,368 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agent instructions
Reference material for building and shipping a mobile app on Supabase. Each file below is self-contained guidance; read the one that matches the task rather than all of them.
These are written around bugs that actually shipped, so each leads with a symptom. If you are debugging something that matches a symptom in the table, read that file before continuing.
When to read what
| Read | When the task involves |
skills/app-flow-design/SKILL.md | Planning onboarding or any multi-step flow; adding a screen that loads data; writing a conditional that picks between screens; a screen showing the wrong thing briefly, showing first-run prompts to returning users, or trapping someone with no way back |
skills/supabase-rls-data-model/SKILL.md | Creating or changing a Supabase table; writing or debugging an RLS policy; adding an invite, pairing or join flow; a query returning nothing for one user and working for another |
skills/supabase-edge-functions/SKILL.md | Anything under supabase/functions/; choosing between the anon and service-role client; secrets or Vault; wiring a Postgres trigger to a function; a function returning 200 without doing its job; a deployed function behaving like older code |
skills/supabase-auth-swift/SKILL.md | Sign-up, sign-in or session handling in SwiftUI; OAuth redirect schemes; what the app does on launch before it knows who the user is; intermittent 401 or cancellation errors just after launch |
skills/ios-app-store-release/SKILL.md | Xcode or XcodeGen project config; preparing a build for upload; anything in App Store Connect; store screenshots; TestFlight; diagnosing a rejection |
Each directory may also hold references/ (read when the SKILL.md
points at them) and scripts/ (run rather than read).
Two rules worth applying everywhere
Design the flow before the schema. The gates in a flow spec are the questions the backend must be able to answer. Building tables first produces predicates shaped by whatever the schema made convenient, which is the origin of the worst bug these notes describe.
Prefer a precise sentence to a convenient condition. partnership
!= nil and "they have actually joined someone" are not the same claim.
When a gate is named after the sentence it asserts, the wrong predicate
becomes visible on sight.
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 28,693 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository10 other files it gives its agents.
- .cursor/rules/app-flow-design.mdcCursor rule
- .cursor/rules/ios-app-store-release.mdcCursor rule
- .cursor/rules/supabase-auth-swift.mdcCursor rule
- .cursor/rules/supabase-edge-functions.mdcCursor rule
- .cursor/rules/supabase-rls-data-model.mdcCursor rule
- app-flow-designSkill
- ios-app-store-releaseSkill
- supabase-auth-swiftSkill
- supabase-edge-functionsSkill
- supabase-rls-data-modelSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.