---
name: repo-baseline
description: >-
  Use when creating a new repository, or auditing an existing repo's hygiene
  and developer tooling. What every repo carries before its first feature:
  hygiene files, secret scanning in pre-commit and CI, one-command onboarding,
  operational script conventions, workflow lifecycle and archiving, in-repo AI
  agent definitions with allowlisted tools and repo-relative paths, declaring
  architecture-standards adoption via `.claude/settings.json` rather than
  relying on memory, and documentation staleness rules.
---

# Repository baseline and developer tooling

**Read [`references/REPO-BASELINE.md`](references/REPO-BASELINE.md) before applying any of this.**
That file is the standard; everything below it is a summary to help you decide
whether this skill applies and to check your work afterwards.

## What this standard covers

- The baseline
- Secret hygiene
- One-command onboarding
- Operational scripts
- Scripts that onboard the *product's* user
- A dependency budget, declared in public
- Research artifacts live in the repository
- Workflow lifecycle
- AI agent definitions live in the repo
- Standards adoption is declared, not remembered
- Documentation staleness

## Checklist

- [ ] `CODEOWNERS`, dependency-update automation, `.editorconfig`, central package management, PR/issue templates, real `.gitattributes`, exclusion-based `.dockerignore`
- [ ] Secret scanning pre-commit + CI; local scripts read secrets from gitignored `.env` with a committed example file; rotation before history-scrubbing
- [ ] One-command interactive setup: prerequisites, secret store init, generated mandatory secret, labeled optional integrations; error-text-keyed troubleshooting table; secret-flow documented
- [ ] Where the product's own users configure something: an enable/disable script that documents the vendor's dialect, cleans up after older versions of itself, and ends with a real-traffic smoke test
- [ ] Per-project dependency counts published; hand-rolling justified only by a small, stable, versioned protocol subset against a high dependency cost, and written forward-compatible
- [ ] Algorithms whose correctness is arguable carry an in-repo reference notebook, proposals with impl status / acceptance criterion / code entry point, shared numbering, and PDFs built in CI rather than committed
- [ ] Runbook scripts numbered + delegating; hand-off token files with documented resolution order; self-sufficient scripts; scripts README with variable tiers; CI jobs mirrored locally; destroy lists carry legacy names
- [ ] Retired workflows archived to `workflows-archive/`, never comment-disabled
- [ ] AI agent definitions in-repo: allowlisted tools, example-bearing descriptions, committed memory policy, repo-relative paths
- [ ] If this repo is meant to conform to `architecture-standards`: `.claude/settings.json` declares the marketplace and enables `architecture-core`
- [ ] README claims verified in review; one named source of truth per environment variable

---

Generated from [`docs/guides/REPO-BASELINE.md`](https://github.com/konradcinkusz/architecture-standards/blob/main/docs/guides/REPO-BASELINE.md) by `scripts/build-marketplace.mjs`. Do not edit this file: change the source document, or its entry in `catalog/marketplace.catalog.json`, and re-run the generator.