メニュー
Bootstrap or maintain a repo-specific Copilot, Codex, and Claude agent stack by inventorying docs, commands, workflows, and existing AI assets before scaffolding managed guidance.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Process and manipulate images using ImageMagick. Supports resizing, format conversion, batch processing, and retrieving image metadata. Use when working with images, creating thumbnails, resizing wallpapers, or performing batch image operations.
Framework for building LLM-powered applications with agents, chains, and RAG. Supports multiple providers (OpenAI, Anthropic, Google), ReAct agents, tool calling, memory management, and vector store retrieval. Use for building chatbots, question-answering systems, autonomous agents, or RAG applications.
Analyze a GitHub user or developer's visible activity across pull requests, issues, commits, reviews, repositories, and discussions. Use when asked to summarize a developer, create a contributor profile, generate a weekly or monthly GitHub roundup, analyze work patterns, prepare GitHub evidence for self-review or promotion, or tell the story of a user's GitHub work.
Bootstrap or maintain a repo-specific Copilot, Codex, and Claude agent stack by inventorying docs, commands, workflows, and existing AI assets before scaffolding managed guidance.
This skill should be used when the user asks to "start an LLM project", "design batch pipeline", "evaluate task-model fit", "structure agent project", or mentions pipeline architecture, agent-assisted development, cost estimation, or choosing between LLM and traditional approaches.
Implement LangGraph error handling with current v1 patterns. Use when users need to classify failures, add RetryPolicy for transient issues, build LLM recovery loops with Command routing, add human-in-the-loop with interrupt()/resume, handle ToolNode errors, or choose a safe strategy between retry, recovery, and escalation.
| name | repo-agent-bootstrap |
| description | Bootstrap or maintain a repo-specific Copilot, Codex, and Claude agent stack by inventorying docs, commands, workflows, and existing AI assets before scaffolding managed guidance. |
This is the Codex-discoverable mirror of the canonical repo-local skill at:
../../.github/skills/repo-agent-bootstrap/
Use the canonical repo-local resources for scripts, references, and templates:
../../.github/skills/repo-agent-bootstrap/scripts/../../.github/skills/repo-agent-bootstrap/references/../../.github/skills/repo-agent-bootstrap/assets/templates/stack/python .github/skills/repo-agent-bootstrap/scripts/inventory_repo.py --repo-root .
python .github/skills/repo-agent-bootstrap/scripts/scaffold_agent_stack.py --repo-root . --generated-on 2026-04-15
python .github/skills/repo-agent-bootstrap/scripts/validate_agent_stack.py --repo-root .
custom-agent for delegation.End with:
From this repository, you can publish the canonical repo-local skill to your shared global skills folder with:
.\scripts\install-global-skill.ps1 -SkillName repo-agent-bootstrap