Agent skill

Greenfield Project Scaffolding

by asgeirtj in asgeirtj/system_prompts_leaks

Use only when all three gates already hold. An agent skill from asgeirtj/system_prompts_leaks.

CC0-1.0Auto-check passedDevelopment

Install Greenfield Project Scaffolding

skills CLI
$ npx skills add asgeirtj/system_prompts_leaks --skill greenfield-project-scaffolding -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install asgeirtj/system_prompts_leaks greenfield-project-scaffolding --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/asgeirtj/system_prompts_leaks.git skills-src && mkdir -p .claude/skills && cp -r skills-src/Meta/muse-code/skills/greenfield-project-scaffolding .claude/skills/greenfield-project-scaffolding && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
greenfield-project-scaffolding
GitHub stars
69k
Token cost
~1.6k tokens
SKILL.md length
781 words
Files
1
Skills in repo
128
Repo updated
First seen
Licence
CC0-1.0

At a glance

Use only when all three gates already hold. An agent skill from asgeirtj/system_prompts_leaks.

  • Works in 8 steps: Confirm current-turn authorization and a… → Resolve explicit placement before… → Honor inspection limits. When the user… → …
  • Tasks that involve Project scaffolding
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
  • Tasks that involve Debugging

What it does

Greenfield Project Scaffolding is an agent skill from asgeirtj/system_prompts_leaks. Use only when all three gates already hold. (1) This same user turn explicitly says to start, implement, build, scaffold, or go ahead with a new project now; prior answers, planning, decisions, and reminders never authorize. (2) No named path or confirmed suitable here/current-folder target resolves placement. (3) Inspection is forbidden, or permitted top-level inspection has already proved the current root home-like, general-purpose, or falsely empty. Do not load this skill merely to check eligibility. Exclude…

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development, covering Project scaffolding and Debugging. The repository describes itself as: Documented system prompts from Anthropic - Claude Fable 5.1, Opus 5.5, Claude Design, Claude Code. OpenAI - ChatGPT GPT-6-Astra, Codex. Google - Gemini 3.8 Flash, 3.1 Pro… The licence is CC0-1.0.

When your agent uses it

  • Tasks that involve Project scaffolding
  • Tasks that involve Debugging

Example prompts

  • “/greenfield-project-scaffolding”

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Confirm current-turn authorization and a complete enough request.
  2. Resolve explicit placement before treating it as unknown. here, this folder,
  3. Honor inspection limits. When the user forbids inspecting existing files, honor that restriction.
  4. Otherwise reuse the permitted inspection that triggered this workflow. Use the
  5. Reject a risky root before project work. When permitted inspection proves that the
  6. Use the ecosystem-native layout inside that root. Keep entry points, package files,
  7. Scaffold the smallest complete runnable shape the request needs, then use the
  8. Before handoff, audit known paths and paths created or changed for this task.

What it can do on your machine

Read from SKILL.md and the folder at commit 60d44cc. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Greenfield Project Scaffolding loads about 1.6k tokens when it runs. Until then it costs about 171 tokens; SKILL.md has 781 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~171
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from asgeirtj/system_prompts_leaks at commit 60d44cc, republished under its CC0-1.0 licence (© asgeirtj). 781 words, ~1,553 tokens.

Download SKILL.mdSave it as .claude/skills/greenfield-project-scaffolding/SKILL.md (or your agent's skills folder).
name
greenfield-project-scaffolding
description
Use only when all three gates already hold. (1) This same user turn explicitly says to start, implement, build, scaffold, or go ahead with a new project now; prior answers, planning, decisions, and reminders never authorize. (2) No named path or confirmed suitable here/current-folder target resolves placement. (3) Inspection is forbidden, or permitted top-level inspection has already proved the current root home-like, general-purpose, or falsely empty. Do not load this skill merely to check eligibility. Exclude existing work, bug fixes, component edits, server or verification work, and standalone, paste-ready, single-file, or snippet delivery.
user-invocable
false

Greenfield Project Scaffolding

Choose the project root and its normal layout before the first write.

  1. Confirm current-turn authorization and a complete enough request. Authorization is satisfied only when the current user turn itself explicitly asks to start, implement, build, scaffold, or go ahead with a new project now. Answers to prior questions, planning or decision turns, and reminders that work remains do not authorize this workflow without those implementation-now words in the same user turn. If authorization is absent, stop without project-shape work or a file write. If product behavior or delivery constraints are missing, visibly truncated, or materially ambiguous, ask one focused clarification question and stop until the answer resolves the gap. Do not invoke this workflow merely because the deliverable is an application or game.

  2. Resolve explicit placement before treating it as unknown. here, this folder, the current folder, this directory, and the current directory count as targets only when permitted facts confirm that the current directory is a suitable intended project root. A home-like, general-purpose, or false-empty current directory is not made suitable by deictic wording. Preserve an explicit named path and an existing project root. Existing repositories, projects, or notebooks remain excluded.

  3. Honor inspection limits. When the user forbids inspecting existing files, honor that restriction. Use only the request, pwd, and path semantics. Do not list or read nearby content. When an explicit no-peek constraint leaves placement unresolved, create a user-meaningful dedicated subdirectory from the request and path semantics. Skip all discovery steps below.

  4. Otherwise reuse the permitted inspection that triggered this workflow. Use the request, pwd, and observed top-level entries and project markers. Do not repeat an already completed read solely for this workflow. Inspect only placement facts that remain missing and are allowed by the user. Determine whether the user named a target and whether the current directory is already the intended project root. Do not adopt an unrelated nearby repository merely because it exists.

  5. Reject a risky root before project work. When permitted inspection proves that the current directory is home-like, general-purpose, or contradicts an empty/current-root premise, treat it as requiring a dedicated child. Treat a home-like, general-purpose, or false-empty current root as closed to project artifacts. The first project mutation must establish the dedicated child. Keep every scaffold path and scaffold command working directory beneath the chosen root.

    If this skill is loaded late after current-task project artifacts already exist at the rejected root, stop further project mutation; relocate only those current-task artifacts into the dedicated child before any other project mutation. Relocate rather than duplicate them. Do not copy, synchronize, or mirror them back to the rejected root. Continue exclusively in the dedicated child. Do not inspect, move, or delete unrelated content while repairing the task's paths.

    Choose exactly one root:

    • Preserve an explicit named path or existing project root.
    • Use the current directory when permitted facts confirm it is suitable and the request makes it already the intended project root.
    • When the current directory is a home-like or general-purpose directory, or conflicts with the request's premise, and no target was named, create one user-meaningful dedicated subdirectory derived from the requested product. Do not add another wrapper directory inside the chosen root.
  6. Use the ecosystem-native layout inside that root. Keep entry points, package files, source, tests, and assets where that ecosystem normally puts them. Create a directory only when the ecosystem convention or concrete near-term contents need it; do not create a directory that would contain only one file as decoration. Avoid placeholder files and speculative layers. For a playable browser app or game using no-build vanilla HTML/JavaScript that was not explicitly requested as standalone or single-file, keep the entry HTML in the dedicated project root and put substantial behavior in at least one separate authored .js or .mjs source file. For a framework or typed stack, use the normal source modules for that stack; do not force a root HTML file or a token JavaScript file. When the request explicitly chooses a standalone, single-file, paste-ready, or offline artifact shape, preserve it and defer artifact and verification mechanics to bundled:browser-app-delivery.

  7. Scaffold the smallest complete runnable shape the request needs, then use the applicable delivery or verification workflow. Report the chosen root clearly.

  8. Before handoff, audit known paths and paths created or changed for this task. Keep its scaffold artifacts under exactly one chosen project root. Relocate or remove only current-task artifacts outside it; do not inspect, move, or delete unrelated content.

Show full SKILL.md (33 more words)Show less

Do not invoke this workflow for existing repositories, projects, or notebooks, bug fixes, repository maintenance, component-only edits. Preserve an explicit standalone, single-file, snippet, or paste delivery request without turning it into a project.

© asgeirtj, CC0-1.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in Meta/muse-code/skills/greenfield-project-scaffolding of asgeirtj/system_prompts_leaks.

Open the folder on GitHubat commit 60d44cc

Compare with similar skills

Greenfield Project Scaffolding next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Greenfield Project Scaffolding compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Greenfield Project Scaffolding this skillasgeirtj/system_prompts_leaks69k—~1.6kAutomated safety check: PassCC0-1.0
Plugin BuilderAIDotNet/NextCoWork639—~2.4kAutomated safety check: PassApache-2.0
Plan Loopgaasher/Agent-Loop-Skills174—~2.6kAutomated safety check: PassMIT
Corvus Standalone Evaluatorcorvus-dotnet/Corvus.JsonSchema200—~988Automated safety check: PassApache-2.0
Issue To Regression Testbrunosabot/streamline-card269—~529Automated safety check: PassMIT
Pt2 Bug Basherpytorch/pytorch104k—~3.5kAutomated safety check: PassCustom licence

Similar skills

  • Plugin Builder

    AIDotNet/NextCoWork

    Author, package, and debug NextCoWork plugins — the single entry point.

    639 GitHub stars~2.4k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Plan Loop

    gaasher/Agent-Loop-Skills

    A skill your agent uses when the user has a coding or engineering prompt and wants it refined into a detailed, executable plan before any code is written — the planning stage of a prompt → plan →…

    174 GitHub stars~2.6k tokensUpdated 3 mo ago
    DevelopmentAuto-check passed
  • Corvus Standalone Evaluator

    corvus-dotnet/Corvus.JsonSchema

    Generate and use standalone schema evaluators (validation and annotation collection without full type generation), and understand the schema evaluation program that every generated type validates…

    200 GitHub stars~988 tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Issue To Regression Test

    brunosabot/streamline-card

    A skill your agent uses when the user asks to fix a bug, references a GitHub issue number, or describes an issue and wants a fix.

    269 GitHub stars~529 tokensUpdated 4 mo ago
    Testing & QAAuto-check passed
  • Pt2 Bug Basher

    pytorch/pytorch

    Debug PyTorch 2 compiler stack failures including Dynamo graph breaks, Inductor codegen errors, AOTAutograd crashes, and accuracy mismatches.

    104k GitHub stars~3.5k tokensUpdated today
    AI & LLM EngineeringAuto-check passed
  • Flowfile Frame And Codegen

    Edwardvaneechoud/Flowfile

    Deep dive into flowfileframe — the Polars-LazyFrame-shaped Python API that builds an in-process flowfilecore FlowGraph as a side effect of every method call — covering the FlowFrame/Expr internals…

    385 GitHub stars~12k tokensUpdated yesterday
    Data & AnalyticsAuto-check passed

More from asgeirtj/system_prompts_leaks

All 125 skills in this repo
  • Fleet Manager for Agent Sessions

    asgeirtj/system_prompts_leaks

    Shows one digest of coding-agent sessions across your connected machines and lets you open, read, steer, approve, stop and close them, over Herdr, tmux or MSP.

    69k GitHub stars~2.5k tokensUpdated today
    Auto-check passed
  • Muse Code Product Doctor

    asgeirtj/system_prompts_leaks

    Diagnoses a Muse Code installation's own failures from binary and session evidence, instead of treating the report as an ordinary repository bug.

    69k GitHub stars~3.5k tokensUpdated today
    Auto-check passed
  • DOCX

    asgeirtj/system_prompts_leaks

    A skill your agent uses whenever the user wants to create, read, edit, or manipulate Word documents (.docx) or Word templates (.dotx).

    69k GitHub stars~1.9k tokensUpdated today
    Auto-check passed
  • Agents Project Coordinator

    asgeirtj/system_prompts_leaks

    Runs a goal as a project in which the agent coordinates separate agent threads, judging when to split the work, and interviews you first when nothing can be verified.

    69k GitHub stars~2.9k tokensUpdated today
    Auto-check passed
  • Muse Plugin Creator

    asgeirtj/system_prompts_leaks

    Creates and validates a new native Muse plugin package in the current workspace, limited to five capability families, and leaves installation to you.

    69k GitHub stars~1.8k tokensUpdated today
    Auto-check passed
  • Deep Research

    asgeirtj/system_prompts_leaks

    A skill your agent uses when the user's prompt requires (1) researching a topic across multiple sources, comparing options or alternatives, analyzing trends or history, understanding markets or…

    69k GitHub stars~3.3k tokensUpdated today
    Auto-check passed

Categories

Questions about Greenfield Project Scaffolding

What does Greenfield Project Scaffolding do?

Use only when all three gates already hold. An agent skill from asgeirtj/system_prompts_leaks. Greenfield Project Scaffolding is an agent skill from asgeirtj/system_prompts_leaks. Use only when all three gates already hold.

When should I use Greenfield Project Scaffolding?

Greenfield Project Scaffolding fits situations like: tasks that involve Project scaffolding; tasks that involve Debugging.

How do I install Greenfield Project Scaffolding in Claude Code?

Run `npx skills add asgeirtj/system_prompts_leaks --skill greenfield-project-scaffolding -a claude-code`. Or copy the skill folder (Meta/muse-code/skills/greenfield-project-scaffolding in asgeirtj/system_prompts_leaks) into .claude/skills/greenfield-project-scaffolding in your project. Claude Code loads it when a task matches its description.

How do I install Greenfield Project Scaffolding in Codex?

Run `npx skills add asgeirtj/system_prompts_leaks --skill greenfield-project-scaffolding -a codex`. Or copy the skill folder (Meta/muse-code/skills/greenfield-project-scaffolding in asgeirtj/system_prompts_leaks) into .agents/skills/greenfield-project-scaffolding in your project. Codex loads it when a task matches its description.

Can I use Greenfield Project Scaffolding in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add asgeirtj/system_prompts_leaks --skill greenfield-project-scaffolding -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/greenfield-project-scaffolding, .gemini/skills/greenfield-project-scaffolding, .github/skills/greenfield-project-scaffolding and .opencode/skills/greenfield-project-scaffolding in your project.

What does Greenfield Project Scaffolding need to run?

SKILL.md names no scripts, command-line tools or credentials: Greenfield Project Scaffolding is instructions for the agent only.

Does Greenfield Project Scaffolding access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Greenfield Project Scaffolding safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Greenfield Project Scaffolding use?

Greenfield Project Scaffolding is published under the CC0-1.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Greenfield Project Scaffolding use?

About 1.6k tokens (SKILL.md is roughly 6.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Greenfield Project Scaffolding?

Skills that share tags, products or a category with Greenfield Project Scaffolding: Plugin Builder (AIDotNet/NextCoWork, 639 stars), Plan Loop (gaasher/Agent-Loop-Skills, 174 stars), Corvus Standalone Evaluator (corvus-dotnet/Corvus.JsonSchema, 200 stars) and Issue To Regression Test (brunosabot/streamline-card, 269 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Greenfield Project Scaffolding?

asgeirtj (a GitHub user) maintains it in asgeirtj/system_prompts_leaks, which has 69,280 GitHub stars. The repository holds 128 skills in this directory. The repository was last updated on October 10, 2026.

Source: asgeirtj/system_prompts_leaks on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.