# Repo Baseline

> 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.

- Skill: `konradcinkusz/repo-baseline` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add konradcinkusz/repo-baseline`
- Raw SKILL.md: https://api.skillmd.com/api/skills/konradcinkusz/repo-baseline/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: konradcinkusz (https://skillmd.com/u/konradcinkusz)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/konradcinkusz/repo-baseline

---


# 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.