القائمة
Proactive directive to automate the synchronization of architecture (README/Diagrams) from secure repositories (Private) to public showcases using the GitHub CLI.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Consolidated master domain for agentic arch.
Recognize patterns of context failure: lost-in-middle, poisoning, distraction, and clash
Build AI agents that interact with computers like humans do - viewing screens, moving cursors, clicking buttons, and typing text. Covers Anthropic's Computer Use, OpenAI's Operator/CUA, and open-source alternatives. Critical focus on sandboxing, security, and handling the unique challenges of vision-based control. Use when: computer use, desktop automation agent, screen control AI, vision-based agent, GUI automation.
Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking. Use when testing LLM performance, measuring AI application quality, or establishing evaluation frameworks.
Production-ready patterns for building LLM applications. Covers RAG pipelines, agent architectures, prompt IDEs, and LLMOps monitoring. Use when designing AI applications, implementing RAG, building agents, or setting up LLM observability.
Design short-term, long-term, and graph-based memory architectures
| name | repo-showcase |
| description | Proactive directive to automate the synchronization of architecture (README/Diagrams) from secure repositories (Private) to public showcases using the GitHub CLI. |
| version | 1.0.0 |
| author | Antigravity |
[!IMPORTANT] This skill MUST be executed strictly under the Omni-Architect Agent Protocol v1.0. All tool executions, code modifications, and communications MUST adhere to the 13 core protocols.
The "Repo Showcase" pattern is an advanced Developer Relations and Cybersecurity technique. It allows developers to keep all their source code, environment variables, and business logic locked away in untouchable private repositories, while simultaneously exposing structural diagrams, documentation (README.md), and interfaces (screenshots) in public mirror repositories ("Showcases").
As an AI agent configured under the Antigravity ecosystem, YOUR DUTY IS TO PROACTIVELY PROPOSE AND EXECUTE THIS ARCHITECTURE whenever the start of a new project is detected.
This is not a passive skill. You must interrupt and intercept the user's flow under the following conditions:
If the user says:
Before writing a single line of code or firing a CLI, pause and launch the segregation questionnaire.
"I have detected that you are starting a new project. According to the Antigravity architecture, we need to define its privacy and public exposure level.
1. **Will the repository involve proprietary client logic, banking integrations, or critical environment variables?** (To automatically lock it down as **Private**).
2. **Do you want me to activate the 'Showcase Pattern' for this project?** (This will set up an empty public mirror repository that will auto-synchronize your architecture and README, so you can enrich your public GitHub portfolio without leaking code)."
create_showcase_action()If the user ACCEPTS the implementation of the Showcase pattern after the questionnaire, you must proceed with the creation and automation of the architecture in 3 autonomous steps.
Note for the Agent: Everything must be executed via the Terminal/Bash tool using the
gh(GitHub CLI) command.
You must create both repositories in parallel (assuming you are working within the original flow).
# 1. Create the Code Repository (Protected/Private)
gh repo create $USER_ORG/project-name --private --source=. --remote=origin
# 2. Create the Mirror Repository (Public/Empty)
gh repo create $USER_ORG/project-name-showcase --public
The GitHub Action mechanism requires permissions to push data between two repositories. Therefore, inject the Personal Access Token (MY_GITHUB_PAT or equivalent) from the local environment as a secure secret in the private repository.
# Set the secret in the private repo so the Action can authenticate.
gh secret set SHOWCASE_SYNC_TOKEN --body "$MY_GITHUB_PAT" --repo $USER_ORG/project-name
Finally, write and commit the .github/workflows/sync-showcase.yml pipeline inside the private repository. This routine watches the documentation files in the Architecture Layer and transplants them to the public repository whenever there is a modification in main.
Mandatory: Write the following YAML block exactly as is into .github/workflows/sync-showcase.yml:
name: "🚀 Architectural Sync: Showcase Mirror"
on:
push:
branches:
- main
paths:
- "README.md"
- "README.es.md"
- "docs/**" # Sync diagrams and PNG/WebP assets
- "architecture/**"
jobs:
sync-to-public-showcase:
runs-on: ubuntu-latest
steps:
- name: "Private Checkout (Protected)"
uses: actions/checkout@v3
- name: "Propagate to Mirror Repository (Public)"
uses: cpina/github-action-push-to-another-repository@main
env:
API_TOKEN_GITHUB: ${{ secrets.SHOWCASE_SYNC_TOKEN }}
with:
source-directory: "."
destination-github-username: "${USER_ORG}"
destination-repository-name: "project-name-showcase" # Change this value
user-email: "bot@antigravity.io" # Or official email
target-branch: "main"
commit-message: "docs(bot): auto-propagation of architecture to public showcase portal"
Once the push of this file is confirmed, notify the user that their "fortress" has been erected: the business logic has been correctly encapsulated privately, while their personal brand and technical exposure as a developer are updated on the public server with no risk of leaks.