Gh Bootstrap

> In the commands below, ` is this skill's base directory, announced when the skill loads. Substitute the literal path; it is not an environment variable. The runtime script ghbootstrapruntime.py` is the only execution engine — never hand-roll the detect / fetch / render / validate steps.

1. Ask the user for communication language and template language preference, plus the desired mode: quick (sensible defaults, minimal interaction), custom (choose specific templates), or full (all available templates).
2. Detect languages, frameworks, and existing GitHub files with python "/scripts/gh_bootstrap_runtime.py" detect "$ARGUMENTS" before planning. If detection returns no languages (empty project or unrecognized stack), ask the user to specify the primary language manually instead of guessing.
3. Call AskUserQuestion to collect necessary configuration variables (project name, GitHub owner/repo, author, license, and any component-specific values).
4. Scan for existing config files and plan conflict resolution. Never overwrite existing files without explicit user approval.
5. Read specs/template-catalog.md to map required files to repository URLs, then use python "/scripts/gh_bootstrap_runtime.py" fetch-template ... to download the chosen template sources. If a template download fails (network error, 404, timeout), report which template failed and offer to skip it or retry.
6. Render files with python "/scripts/gh_bootstrap_runtime.py" render-template ... so placeholder replacement is deterministic.
7. Validate the generated tree with python "/scripts/gh_bootstrap_runtime.py" validate-tree and refuse completion if unreplaced {{placeholders}} remain.

Mandatory Rules

• NEVER write config files from memory; MUST render from downloaded templates.
• ALWAYS replace {{variable}} placeholders (e.g. {{projectName}}); NEVER alter GitHub Actions expressions (${{ github.* }}, ${{ secrets.* }}) or action version pins (e.g. @v4). The renderer only substitutes {{...}}.
• Output targets .github/workflows/, issue templates, PR templates, and baseline community files.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: bahayonghang
• Source: bahayonghang/my-ai-cli-toolkit
• License: MIT

Install and usage instructions live in the source repository linked above.