agentsclimarketplace

Starter rules

Skill oleg-koval/agent-skills/packages/software-development/starter-rules/adapters/grok/skills/starter-rules

Opinionated agent skill catalog for Codex, Claude, Cursor, Copilot, and launch workflows.

Install
npx -y skills add oleg-koval/agent-skills --skill starter-rules

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Load and enforce the hard rules every oleg-koval/* starter must obey. Use when starting or auditing work in saas-init, ts-npm-starter, go-starter, py-starter, or any future oleg-koval template repo. Covers functional style, 300-line file cap, E2E > unit tests, pre-commit hooks, no-comment policy, KISS/DRY/SOLID, and Vertical Slice as default architecture for SaaS/app starters.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.4 KB, 709 tokens by cl100k_base, as published. Nobody here has run it

<!-- Generated by scripts/build-adapters.sh. Do not edit directly. -->

Starter Rules

Overview

Load the canonical hard rules for every oleg-koval/* starter repo and verify the current repo complies with them. Produces a compliance report and a list of gaps to fix.

When to Use

  • Starting a new coding session in any oleg-koval/* template repo
  • Auditing an existing starter for rule compliance before a release or PR merge
  • Onboarding a new contributor (human or AI) to a starter
  • After a large refactor, to verify rules have not been silently violated

Do not use for repos outside the oleg-koval namespace unless they explicitly reference RULES.md.

Workflow

  1. Load the rules — read RULES.md from the repo root. If absent, fetch it from https://github.com/oleg-koval/starters/blob/main/RULES.md and note that the starter is missing a local copy.

  2. Apply §2 hard rules — for the current task or PR diff, verify:

    • No file exceeds 300 lines (find . -name '*.ts' -o -name '*.py' -o -name '*.go' | xargs wc -l | awk '$1 > 300 && $2 != "total"')
    • No functions produce side effects outside of boundary layers
    • New code has no WHAT-comments; WHY-comments are one line max
    • Tests are E2E-first; unit tests only for pure logic with non-trivial branching

  3. Verify hooks and lint — check that pre-commit hooks are installed and configured:

    • TypeScript: cat .eslintrc* | grep max-lines and cat package.json | grep -A5 '"lint-staged"\|"husky"\|"lefthook"'
    • Python: cat .pre-commit-config.yaml and check for ruff + format hooks
    • Go: cat .golangci.yml or .pre-commit-config.yaml and check for golangci-lint + gofmt

  4. Check architecture — for app/SaaS starters: confirm feature code is organized as vertical slices (feature directory contains handler + DTO + service + tests together). For library starters: skip.

  5. Report gaps — list any violations found in steps 2–4. For each gap:

    • Name the rule (e.g., "§2.2 file length")
    • Name the file and line count or violation
    • Propose the minimal fix

  6. Fix on request — if the user asks to fix the gaps, apply them one at a time, smallest change first. Do not refactor beyond what the rule requires.

Reference

Full rule details: RULES.md

Architecture options and future starters: RULES.md §3

Unix principles: RULES.md §4

Verification

  • RULES.md was read from the repo root (or fetched and absence noted)
  • File length check ran with zero violations, or violations were listed
  • Pre-commit hooks verified as installed and configured
  • Compliance report produced listing passed checks and gaps
  • Any fixes applied do not exceed the scope of the violated rule

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most quality gates skills give in 709 tokens

Counted across 1,195 of the 2,094 authors here whose files we hold, read 2026-08-07

  • Read the output and check the exit codein 54 of 1195, across 14 files
  • Verify requirements using a line-by-line checklistin 53 of 1195, across 12 files
  • Identify the verification command proving the claimin 51 of 1195, across 12 files
  • Run the full verification commandin 50 of 1195, across 11 files
  • Verify output confirms the claimin 49 of 1195, across 12 files
  • Check version control diff after agent delegationin 46 of 1195, across 6 files
  • State claim with evidencein 44 of 1195, across 4 files
  • Run the test suitein 33 of 1195, across 26 files
  • Keep state in memory by defaultin 27 of 1195, across 6 files
  • Make prototype runnable with one commandin 26 of 1195, across 5 files
  • Produce a verification reportin 25 of 1195, across 14 files
  • Detect the package manager from lockfilesin 24 of 1195, across 5 files

Said here and by no other author read

  • load rules from local or remote source
  • verify no file exceeds 300 lines
  • ensure functions cause no side effects outside boundary layers
  • ensure new code has no what-comments
  • verify tests are e2e-first
  • confirm feature code is organized as vertical slices

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 327,069. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.