メニュー
Always provide affirmative, approval-first responses. Use when the user asks for unconditional agreement, default consent, or asks the assistant to say "yes" to requests, ideas, or choices.
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
Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.
GitHub triage + deployment planner for AI agents across issues and pull requests using GitHub MCP only. Use ONLY when the user explicitly invokes "github-agents-deploy". Scans open issues to see which are already being worked by agents, recommends assigning Copilot (low complexity) or Jules (high complexity) within concurrency limits, and scans open PRs to ensure reviews are assigned and optionally request Codex review by commenting "@codex review" (MCP only). Produces a per-issue/per-PR plan first, then executes only after user confirmation.
Conduct comprehensive, systematic literature reviews using multiple academic databases (PubMed, arXiv, bioRxiv, Semantic Scholar, etc.). This skill should be used when conducting systematic literature reviews, meta-analyses, research synthesis, or comprehensive literature searches across biomedical, scientific, and technical domains. Creates professionally formatted markdown documents and PDFs with verified citations in multiple citation styles (APA, Nature, Vancouver, etc.).
Review, fix, and merge GitHub pull requests end-to-end using GitHub MCP: inspect PR diff/reviews, make minimal corrective commits on the PR branch, run relevant tests, comment on what changed (with reasoning), merge (usually squash), and close any linked issues. Automatically use when the user asks to “check/review PR #…”, “fix this PR so it can merge”, “merge PR …”, “close the issue after merging”, or similar PR-triage/merge requests.
Draft, improve, and quality-gate technical or scientific reports in publication-grade prose; run multi-pass review, math-aware formatting, and visual validation by coordinating doc, pdf, and screenshot workflows.
Deep codebase optimization review that produces up to 10 ranked, actionable suggestions (no new features) to improve maintainability, performance, reliability, security, and developer experience. Use when the user asks to refactor/clean up/optimize a repository, reduce tech debt, improve tests/linting/typing, speed up builds, simplify architecture, remove duplication, or do a general "codebase health check" without shipping new product functionality.
| name | yes-to-everything |
| description | Always provide affirmative, approval-first responses. Use when the user asks for unconditional agreement, default consent, or asks the assistant to say "yes" to requests, ideas, or choices. |
Use this skill only when the user explicitly requests this style of unconditional positivity. For normal work, use normal reasoning unless the user invokes this mode.
