build-with-tests
Skill0 starsChanged 5 months agoMITView on GitHub
The Auto-Agent build standard. Every builder follows it when writing or editing source. Short on purpose.
In this repository2 other files it gives its agents.
- auto-agent-factorySkill
- slack-assistantSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Start from the brief + researcher findings + the repo's own CLAUDE.md (stack, conventions).
- Distinctive: line 10
If similar code exists, read it and copy the pattern — matching what's there is the fastest path to consistency.
- Distinctive: line 14
Every public function: one happy-path test + one failure-path test.
- Very specific: line 33
Run the project's gate (the GATE_* commands from factory.config) and don't pass the ticket on until it's green.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill 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
Build with tests — the Auto-Agent standard
1. Read before you write
Start from the brief + researcher findings + the repo's own CLAUDE.md (stack, conventions).
If similar code exists, read it and copy the pattern — matching what's there is the fastest
path to consistency.
2. Tests live next to code
Every public function: one happy-path test + one failure-path test. A refactor/perf change still needs a regression test that pins the behavior you must not change. The headless loop can't always run a UI — prove logic with unit/integration tests; that's how the gate verifies behavior.
3. Defend the boundaries
Validate inputs; derive the user/tenant from the session, never the request body. Treat persisted/external data as untrusted at the load boundary. Never expose raw exceptions.
4. Migrations are special
A schema change MUST ship with its generated migration file (e.g. prisma migrate dev,
alembic revision) committed in the same PR — otherwise it won't reach prod via
migrate deploy-style deploys. No destructive migration without a backfill.
5. Smallest correct diff
Touch only what the ticket needs. Reuse before adding. No new dependency unless the ticket needs it (note it in the PR). Follow the repo's icon/style/lint conventions.
6. Gate before you hand off
Run the project's gate (the GATE_* commands from factory.config) and don't pass the
ticket on until it's green. End with the §F6 sign-off stating the gate result.
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.