Agent skill

Create Feature

by zacharyfmarion in zacharyfmarion/openscad-studio

A skill your agent uses when the user asks to take a feature or bug fix from prompt to implementation, especially prompts like "/create a new feature", "build this feature end-to-end", "take this…

GPL-2.0Auto-check passedAgent Workflows

Install Create Feature

skills CLI
$ npx skills add zacharyfmarion/openscad-studio --skill create-feature -a claude-code

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

GitHub CLI
$ gh skill install zacharyfmarion/openscad-studio create-feature --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/zacharyfmarion/openscad-studio.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/create-feature .claude/skills/create-feature && 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
create-feature
GitHub stars
238
Token cost
~1.5k tokens
SKILL.md length
751 words
Files
2
Skills in repo
2
Repo updated
First seen
Licence
GPL-2.0

At a glance

A skill your agent uses when the user asks to take a feature or bug fix from prompt to implementation, especially prompts like "/create a new feature", "build this feature end-to-end", "take this…

  • Works in 4 steps: AGENTS.md → CLAUDE.md → .github/PULL_REQUEST_TEMPLATE.md → …
  • The user asks to take a feature
  • SKILL.md covers What This Skill Owns, Required Reads, Checkout Readiness and Planning Contract, plus 6 more sections
  • Calls pnpm, bash and gh

What it does

Create Feature is an agent skill from zacharyfmarion/openscad-studio. Use when the user asks to take a feature or bug fix from prompt to implementation, especially prompts like "/create a new feature", "build this feature end-to-end", "take this from plan to PR", or "own this change through validation and handoff". This skill is for repo-local delivery workflows that must create and maintain implementation plans, choose appropriate tests, run deterministic validation, open a draft PR against main, and return the preview deployment URL when one is available.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Agent Workflows, covering Planning, Debugging and Deployment. It works with WebAssembly and macOS. The repository describes itself as: Create 2D and 3D designs with AI. The licence is GPL-2.0.

When your agent uses it

  • The user asks to take a feature
  • Bug fix from prompt to implementation
  • Especially prompts like /create a new feature
  • Build this feature end-to-end

Example prompts

  • “/create a new feature”
  • “build this feature end-to-end”
  • “take this from plan to PR”
  • “/create-feature”

Workflow steps

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

  1. AGENTS.md
  2. CLAUDE.md
  3. .github/PULL_REQUEST_TEMPLATE.md
  4. Any relevant existing file under implementation-plans/

What it can do on your machine

Read from SKILL.md and the folder at commit 285da7d. 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

    Shell commands in SKILL.md call:

    • pnpm
    • bash
    • gh

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

  • Network

    No URLs in SKILL.md. Its commands use pnpm and gh, which can reach the network depending on how they are called.

    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

Create Feature loads about 1.5k tokens when it runs. Until then it costs about 127 tokens; SKILL.md has 751 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~127
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

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 zacharyfmarion/openscad-studio at commit 285da7d, republished under its GPL-2.0 licence (© zacharyfmarion). 751 words, ~1,540 tokens.

Download SKILL.mdSave it as .claude/skills/create-feature/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
create-feature
description
Use when the user asks to take a feature or bug fix from prompt to implementation, especially prompts like "/create a new feature", "build this feature end-to-end", "take this from plan to PR", or "own this change through validation and handoff". This skill is for repo-local delivery workflows that must create and maintain implementation plans, choose appropriate tests, run deterministic validation, open a draft PR against main, and return the preview deployment URL when one is available.

Create Feature

Use this skill when the user wants one agent to carry a repo change through planning, execution, validation, and PR handoff.

What This Skill Owns

  • Create and maintain an implementation plan for non-trivial work.
  • Inspect the current checkout and make sure it is ready before editing.
  • Implement the requested change directly unless a material product decision blocks progress.
  • Add or update tests that match the changed behavior.
  • Run local validation through the shared validation script.
  • Open a draft PR against main.
  • Wait for the PR preview workflow and return both the PR URL and preview URL when a web preview applies.

Required Reads

Before changing code, read these repo guides:

  1. AGENTS.md
  2. CLAUDE.md
  3. .github/PULL_REQUEST_TEMPLATE.md
  4. Any relevant existing file under implementation-plans/

Checkout Readiness

Do not create a worktree in this skill.

Instead:

  1. Inspect checkout state with non-interactive Git commands.
  2. If the repo is in a Git worktree, make sure that worktree is actually ready for development.
  3. If the repo is in a normal checkout, continue in place.
  4. Treat setup as a readiness check, not a separate provisioning workflow.

Default readiness expectations:

  • Ensure dependencies are installed when needed with pnpm install --frozen-lockfile.
  • Ensure Playwright browsers are installed only when web E2E is needed.
  • Ensure Rust/Tauri tooling is present only when the changed scope touches apps/ui/src-tauri or other desktop-only behavior.

Planning Contract

For non-trivial work, derive a concise slug from the task and create implementation-plans/<slug>.md.

Use the repo's established plan format:

  • # <Title>
  • ## Goal
  • ## Approach
  • ## Affected Areas
  • ## Checklist

Keep the checklist current while you work. Mark steps complete as soon as they are actually done.

Do not create an implementation plan for narrow housekeeping work such as typo-only edits, formatting cleanup, or other small maintenance.

Execution Contract

After planning, implement directly unless blocked by a real product ambiguity.

Always:

  • Prefer the smallest change that fully solves the task.
  • Read existing patterns before introducing new ones.
  • Avoid inventing new repo workflow conventions when an existing one can be extended.
  • Keep a running summary of what changed and why for the PR body.

Test Expectations

Choose tests based on the changed behavior:

  • Add or update unit or component tests for changed logic.
  • Add or update formatter regression tests when formatter behavior changes.
  • Add or update Playwright coverage when the change introduces or materially changes a user-facing flow.
  • If no new tests are needed, be ready to justify that in the PR.
Show full SKILL.md (347 more words)Show less

Validation Contract

Use scripts/validate-changes.sh for deterministic command execution.

You are responsible for choosing scopes; the script is responsible for running them consistently.

Available scopes:

  • baseline: pnpm format:check, pnpm lint, pnpm type-check, pnpm test:unit
  • formatter: pnpm test:formatter:ci
  • rust: Rust formatting and static checks under apps/ui/src-tauri
  • e2e-web: pnpm test:e2e:web

Recommended usage:

  • Run pnpm format yourself before validation if the repo needs formatting fixes.
  • Use --fix only when you intentionally want the helper to apply formatting before running the remaining checks.
  • Use explicit scopes when you already know what changed.
  • Use --changed-file <path> inputs when you want the script to infer formatter and rust.
  • Use --dry-run first when you want to confirm the command set before running it.

Examples:

bash
bash scripts/validate-changes.sh --dry-run --changed-file apps/ui/src/components/AiPromptPanel.tsx
bash scripts/validate-changes.sh --scope baseline
bash scripts/validate-changes.sh --fix --scope baseline
bash scripts/validate-changes.sh --changed-file apps/ui/src/utils/formatter/printer.ts
bash scripts/validate-changes.sh --scope baseline --scope e2e-web
bash scripts/validate-changes.sh --scope baseline --scope rust

In your final summary and PR notes, report:

  • Which validations ran
  • Which validations were skipped
  • Why each skipped validation was not necessary

Pull Request Handoff

Unless the user asked otherwise, open a draft PR against main.

Before creating the PR:

  1. Confirm the working tree contains only intended changes.
  2. Fill the PR body using .github/PULL_REQUEST_TEMPLATE.md.
  3. Include the implementation plan path in the PR notes when one was created.
  4. Summarize tests added, validations run, and intentionally skipped checks.

Use gh pr create --draft --base main.

If gh auth or GitHub access is unavailable, stop after local validation and report the exact blocker.

Preview Deployment Handoff

This repo uses a PR preview workflow that deploys web-relevant pull requests to:

  • https://pr-<number>.openscad-studio.pages.dev

After opening the PR:

  1. Wait for the PR preview workflow to finish.
  2. Look for the sticky preview comment or workflow success result.
  3. Return both the PR URL and preview URL when a web preview applies.
  4. If the PR is desktop-only or docs-only, report that no web preview was expected.

Guardrails

  • Do not create or switch worktrees from this skill.
  • Do not skip the implementation plan for non-trivial work.
  • Do not open the PR before required local validation succeeds.
  • Do not claim a preview URL exists before the workflow has published it.
  • Do not target any base branch other than main unless the user explicitly says so.

© zacharyfmarion, GPL-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .agents/skills/create-feature of zacharyfmarion/openscad-studio.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit 285da7d

Compare with similar skills

Create Feature 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.

Create Feature compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Create Feature this skillzacharyfmarion/openscad-studio238—~1.5kAutomated safety check: PassGPL-2.0
Plan Py4vaspvasp-dev/py4vasp100—~2.3kAutomated safety check: PassApache-2.0
OpenWork Release Validationdifferent-ai/openwork24k—~1.7kAutomated safety check: PassCustom licence
SuperpowersPeiiii/nextclaw260—~2.1kAutomated safety check: PassMIT
Plan ModeCodeAlive-AI/ai-driven-development157—~844Automated safety check: PassMIT
Gearcoleco Debuggingdrhelius/Gearcoleco141—~3.5kAutomated safety check: PassGPL-3.0

Similar skills

  • Plan Py4vasp

    vasp-dev/py4vasp

    Plan a py4vasp change as an ordered list of test-first chunks — that chunk list is the plan.

    100 GitHub stars~2.3k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • OpenWork Release Validation

    different-ai/openwork

    Checks a published OpenWork desktop release by booting the released macOS binaries through packaged journeys, verifying signing and updater manifests, and publishing evidence.

    24k GitHub stars~1.7k tokensUpdated today
    DevOps & CloudAuto-check passed
  • Superpowers

    Peiiii/nextclaw

    A skill your agent uses when the user wants a disciplined software development workflow with design-first planning, implementation plans, TDD, systematic debugging, code review, or…

    260 GitHub stars~2.1k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Plan Mode

    CodeAlive-AI/ai-driven-development

    A skill your agent uses when the user asks to plan, design, scope, estimate, or implement a feature, bug fix, refactor, migration, integration, API change, UI change, or other project modification.

    157 GitHub stars~844 tokensUpdated 2 days ago
    Agent WorkflowsAuto-check passed
  • Gearcoleco Debugging

    drhelius/Gearcoleco

    Debug and trace ColecoVision and Super Game Module games using the Gearcoleco emulator MCP server.

    141 GitHub stars~3.5k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Official

    Diagnoses why the Claude in Chrome MCP tools report the browser extension as not connected, with macOS-specific checks and a fix for the Claude.app native host conflict.

    7.4k GitHub starsUsed in 4 repos~2.6k tokens
    Agent WorkflowsAuto-check passed

More from zacharyfmarion/openscad-studio

  • PR

    zacharyfmarion/openscad-studio

    A skill your agent uses when the user wants to open a pull request or finish a branch handoff.

    238 GitHub stars~570 tokensUpdated 1 mo ago
    Auto-check passed

Questions about Create Feature

What does Create Feature do?

A skill your agent uses when the user asks to take a feature or bug fix from prompt to implementation, especially prompts like "/create a new feature", "build this feature end-to-end", "take this…. Create Feature is an agent skill from zacharyfmarion/openscad-studio. Use when the user asks to take a feature or bug fix from prompt to implementation, especially prompts like "/create a new feature", "build this feature end-to-end", "take this from plan to PR", or "own this change through validation and handoff".

When should I use Create Feature?

Create Feature fits situations like: the user asks to take a feature; bug fix from prompt to implementation; especially prompts like /create a new feature; build this feature end-to-end.

How do I install Create Feature in Claude Code?

Run `npx skills add zacharyfmarion/openscad-studio --skill create-feature -a claude-code`. Or copy the skill folder (.agents/skills/create-feature in zacharyfmarion/openscad-studio) into .claude/skills/create-feature in your project. Claude Code loads it when a task matches its description.

How do I install Create Feature in Codex?

Run `npx skills add zacharyfmarion/openscad-studio --skill create-feature -a codex`. Or copy the skill folder (.agents/skills/create-feature in zacharyfmarion/openscad-studio) into .agents/skills/create-feature in your project. Codex loads it when a task matches its description.

Can I use Create Feature 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 zacharyfmarion/openscad-studio --skill create-feature -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/create-feature, .gemini/skills/create-feature, .github/skills/create-feature and .opencode/skills/create-feature in your project.

What does Create Feature need to run?

Going by SKILL.md and its folder, Create Feature needs the command-line tools its instructions call (pnpm, bash and gh).

Does Create Feature access the network?

SKILL.md contains no URLs. Its commands use gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Create Feature 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 Create Feature use?

Create Feature is published under the GPL-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Create Feature use?

About 1.5k 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 Create Feature?

Skills that share tags, products or a category with Create Feature: Plan Py4vasp (vasp-dev/py4vasp, 100 stars), OpenWork Release Validation (different-ai/openwork, 24k stars), Superpowers (Peiiii/nextclaw, 260 stars) and Plan Mode (CodeAlive-AI/ai-driven-development, 157 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Create Feature?

zacharyfmarion (a GitHub user) maintains it in zacharyfmarion/openscad-studio, which has 238 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on August 19, 2026.

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