agentleFS
vaaynevaayne/agent-kit/skills/project-setup/SKILL.md

project-setup

Skill54 starsChanged 27 days agoMITView on GitHub

Set up or audit a repository with V's project baseline. Common: an AGENTS.md index with .agents/{rules,skills}, mise tools and .mise/tasks, prek, dprint, a no-mock test policy with tester-army/e2e, documentation with documd figures and Fumadocs, GitHub commit/issue/PR/CI rules, and PostgreSQL with goose. Stacks: Go (golangci-lint, sqlc, OpenAPI, embedded React UI), React UI (TypeScript 7, Vite+, shadcn on Base UI, Beautiful UI, TanStack Query, anti-slop), and TypeScript full stack (TanStack Start, Effect, @effect/sql-pg). Use when the user starts a new project, asks to apply "best practices" or "项目规范" to a repository, bootstraps AGENTS.md or .agents/, or asks which tools and skills a project should use.

48 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use creative-writing-craft, creative-writing-muse and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Abdullokh-NSkill

    creative-writing-craft

    Справочник ремесла: проза, сцена, стиль, голос. Загружать, когда нужно знание "как писать", а не режим производства.

    Also in 3 projects that use this

  • Abdullokh-NSkill

    creative-writing-muse

    Точка входа в набор скилов для длинной прозы. Включается, когда автор просит спланировать, написать, продолжить, переписать или разобрать художественный текст — роман, повесть, цикл рассказов. Ведёт маршрут от замысла к готовому куску и держит базу знаний проекта.

    Also in 3 projects that use this

  • badgidsSkill

    character-sim

    Pressure-test a character by speaking or reasoning strictly from that character's current voice, goals, relationships, emotional pressure, and knowledge boundary without creating canon by accident.

    Also in 3 projects that use this

  • jpsweeney97Skill

    writing-principles

    Enforces writing principles for instruction documents (SKILL.md, skill supporting files, subagents, CLAUDE.md). Use when writing, reviewing, or editing any file in **/skills/**/*.md, **/agents/*.md, or **/CLAUDE.md. Triggers on skill creation, skill file updates, CLAUDE.md changes, subagent authoring, or instruction document review.

    Also in 3 projects that use this

  • ItsSidharthaSkill

    story-planning

    Collaborative planning skill for a GitHub issue. Takes issue content and codebase context as input. Runs an iterative clarification loop until the user explicitly agrees to a plan, then saves it to docs/<story_number>/plan.md. Used by the user-story-implementation skill — load this skill at the start of Phase 1 and follow it completely before returning to the orchestrator.

    Also in 3 projects that use this

  • badgidsSkill

    story-review

    Diagnose existing prose at a chosen editorial level, from whole-story and developmental problems through scene, line, copy, and proof passes, without rewriting until the diagnosis is accepted.

    Also in 3 projects that use this

  • +4 more from Abdullokh-N

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 4 are distinctive, and 5 are very specific to this project.

  • 0

    Common

    Most of the 40 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Inspect: Read the existing agent instructions, git status, and the manifests (package.json, go.mod, mise.toml, prek.toml, dprint.json).

    line 44
  2. Very specific:

    Lay out instructions: Follow layout. Create the root AGENTS.md index and the .agents/rules/ files.

    line 51
  3. Distinctive:

    Install the common tooling and rules: Follow these for every repository:

    line 57
  4. Distinctive:

    Apply each stack: Read only the matching references:

    line 68
  5. Very specific:

    Install skills: Add each upstream skill to the skills:install task and to .agents/skills/.gitignore, then run mise run skills:install.

    line 78
  6. Very specific:

    Bootstrap: Run mise install, the locked dependency install (mise run install), mise run skills:install, mise run hooks:install, and the …

    line 83
  7. Very specific:

    Verify: After an initial setup, run prek validate-config prek.toml, mise run format, and mise run check.

    line 87
  8. Distinctive:

    Before you add a tool, ask if the repository needs it now.

    line 92
  9. Distinctive:

    When you add a skill, pin the latest release tag.

    line 130

“Common” is measured: kept by at least 6 in 10 of the 40 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Project setup

Apply one baseline to every repository, then add the rules for each stack the repository uses. The target repository owns the result: it pins its own tool versions and installs its own skills. This skill only says what to install and where to get it.

Modes

  • Setup applies the baseline to a new repository, or to an existing one when the user asks for the change.
  • Audit reports the differences between the repository and the baseline, with file references and a proposed change for each. An audit does not install, generate, format, or move files.

State the mode before you start. If the request is not clear, run an audit.

New and existing repositories

The stack choices in this skill (Vite+, TypeScript 7, TanStack Query, Effect, golangci-lint, PostgreSQL) are defaults for new projects. In an existing repository, keep the established stack, package manager, and documentation layout. Report a stack migration as a proposal; do it only when the user puts it in scope. Read the ADRs and data-ownership rules before you interpret leftover configuration.

Procedure

  1. Inspect. Read the existing agent instructions, git status, and the manifests (package.json, go.mod, mise.toml, prek.toml, dprint.json). Identify the stacks. Record any user edits in the working tree and preserve them.

    Complete when you know the stacks and which baseline parts already exist.

  2. Lay out instructions. Follow layout. Create the root AGENTS.md index and the .agents/rules/ files. Keep existing rules that still apply; move them into the matching rule file. When the repository writes documents, also write docs/AGENTS.md from docs-agents and link it from the index.
  3. Install the common tooling and rules. Follow these for every repository:
    • tooling: mise and .mise/tasks, prek, dprint, documd, and e2e.
    • testing: test policy, tiers, coverage, and enforcement.
    • documentation: document layout, the docs/AGENTS.md standard, figures, and Fumadocs for public docs.
    • github: commits, issues, PRs, CI, and Dependabot.
  4. Apply each stack. Read only the matching references:
    • Go service: go.
    • Go service with a web UI: go, typescript, and web.
    • TypeScript full stack: typescript, web, and ts-fullstack.
    • The application owns relational data: database. Persistent data alone does not require PostgreSQL; a file store or an external system of record may own it.
  5. Install skills. Add each upstream skill to the skills:install task and to .agents/skills/.gitignore, then run mise run skills:install. Do not copy skill content into this skill or write it from memory. See skill sources.
  6. Bootstrap. Run mise install, the locked dependency install (mise run install), mise run skills:install, mise run hooks:install, and the browser install for e2e when the repository has one.
  7. Verify. After an initial setup, run prek validate-config prek.toml, mise run format, and mise run check. After an incremental change, run the checks for the changed area, and widen them for shared contracts. Report each check that you did not run and why.

Before you add a tool, ask if the repository needs it now. Skip parts that do not apply. For example, do not add PostgreSQL when the application owns no relational data, do not add e2e to a library without a user interface, and do not add Fumadocs without public documentation.

Skill sources

Skills live in .agents/skills/ at project scope. There are two kinds:

  • Upstream skills come from a GitHub repository (see the table). Do not commit them, because they can be large and the upstream owns them. List each one in .agents/skills/.gitignore, and install it with the skills:install task.
  • Local skills exist only in this repository. Commit them.

The task pins each upstream skill to a tag or a commit SHA, so every checkout gets the same content. --force lets the task run again over an existing install. To update a skill, change its pin and run the task again.

#!/bin/bash
#MISE description="Install the pinned upstream agent skills into .agents/skills"
set -euo pipefail

# Upstream skills are not committed; .agents/skills/.gitignore lists them.
# To update one, change its pin here and run this task again.
install() {
	gh skill install "$1" "$2" --pin "$3" --dir .agents/skills --force
}

install tester-army/e2e e2e <tag-or-sha>
# Upstream skills. `mise run skills:install` installs them at pinned versions.
e2e/

When you add a skill, pin the latest release tag. If the repository has no release, pin the head commit SHA of the default branch. Do not pin a branch name. Review the skill content before you add it to the task. A skill with many files installs slowly, because gh downloads each file separately.

NeedRepositorySkill
Diagrams in Markdownmarkdown-viewer/skillsdocumd-visuals
PostgreSQLneondatabase/postgres-skillspostgres-best-practices
End-to-end teststester-army/e2ee2e
Go codeJetBrains/go-modern-guidelinesuse-modern-go
Oxlint anti-slop installdmmulroy/anti-slopinstall-anti-slop
shadcn componentsshadcn-ui/uishadcn

install-anti-slop is a one-time installer. Run it from its upstream copy during setup; do not keep it in the repository unless the team updates anti-slop often.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.