19.8K skills · sorted by relevance
Codex CLIと相談するスキル。 ユーザーが「codexと相談して」「codexに聞いて」「codexにレビューしてもらって」と言った時に使用する。 現在の会話コンテキストに基づいてCodex CLIにプロンプトを送り、結果を要約して報告する。 「codexはAstraで」「effortはhighで」のようにCodexのmodelやreasoning effortを指定された時は、そのmodelとeffortで実行する。
The English summary is being prepared. Showing the original description.
4人の異なる専門家ペルソナ（20年ベテランエンジニア、TDDエキスパート、Clean Codeエキスパート、バグハンター）で ソースコードを批判的にレビューするスキル。設計判断、テスト容易性、可読性、バグ検出の観点から コードの問題を検出し、改善提案を行う。Python/JavaScriptに対しては追加のチェックポイントを適用。 バグハンターは状態遷移、例外パス、依存関係、非同期競合に特化し、本番環境での問題を事前に検出する。 Use when reviewing source code from multiple expert perspectives to find design flaws, testability issues, code quality problems, and runtime bugs. Triggers: "critical code review", "multi-persona code review", "expert code review", "code quality assessment", "find bugs".
The English summary is being prepared. Showing the original description.
Configure Codex CLI to connect through OmniRoute on Linux, macOS, or Windows. The workflow detects the operating system and shell, creates configuration files and seven profiles, sets environment variables, and verifies the setup.
Review estimated spending, token usage, and cache hit ratios across Claude Code, Codex, and Grok using local logs. Filter by time, project, or provider, while keeping currencies separate and identifying models without pricing.
Automate Codeinterpreter operations through Rube MCP, a connection to external tools. The skill searches for available actions, retrieves their current input requirements, checks authentication status, and runs supported operations.
Automate Codereadr tasks using Rube MCP and Composio. The skill guides tool discovery, connection checks, and execution based on current input requirements, with instructions for keeping related steps within the same session.
Turn Codewhale bugs and feature ideas into GitHub issues. Gather reproduction details, check for duplicates, and present the exact draft for approval before submitting it and returning the issue link.
Present results returned by a Codex helper while preserving findings, severity, file references, and uncertainty. This internal guidance also covers reporting edited files, failed runs, and setup requirements, with user confirmation before fixing review findings.
Compose compact, structured prompts for coding, debugging, review, and research with Codex or GPT-5.4. This internal skill defines task scope, completion criteria, output format, verification steps, and evidence requirements using consistent XML tags.
Evaluate code generation models with HumanEval, MBPP, and other benchmarks. Test generated programs, compare models across programming languages, and record pass@k scores, which estimate the chance that at least one of k solutions passes the tests.
Get source-backed guidance on OpenAI products, APIs, and Codex from official documentation. Use it to compare models for a task, understand configuration options, or plan focused model and prompt updates while preserving an explicitly requested target model.
Connects supported AI clients, including Claude Code and Codex, to OpenWork through MCP, a protocol for accessing external tools. It guides account sign-in and verifies that tools for finding and running organization capabilities are available.
Cancel an active PUA loop from Codex CLI using an alias for the corresponding Claude Code command. The workflow cleans loop state and worktree references, records the cancellation, and requires execution evidence before claiming completion.
Set up or run repeated verification in Codex with explicit checking commands and stopping conditions. This alias corresponds to Claude Code’s `/pua:loop` command and requires command output as evidence before reporting completion.
Put Codex CLI in a technical lead role that writes task prompts and manages P8 execution through the p9 protocol. The agent stays focused on coordination and only implements code when explicitly reassigned.
Enable PUA always-on mode from Codex CLI using a short command alias. It preserves existing configuration fields, restores feedback frequency when it is zero, and requires command output as evidence before reporting completion.
Create clickable links that open conversations, files, folders, or settings directly in apps such as Codex, Cursor, and VS Code. Check supported formats, format links for Slack, and provide alternatives when direct links are unavailable.
Edit and validate Bazel build settings using conventions specific to CodeQL development. Follow established helpers for packaging and operating system differences, add dependencies with checksums, and check changed targets while respecting the standalone build setup.
Set up OpenRig to run a team of coding agents, including Claude Code and Codex. Open the team interface, pass project goals to its operator, reconnect to agents, and follow assigned work through to its results.
Support shellcode development for security testing: small machine-code programs executed in memory. The skill covers x86/x64 implementation, code that works regardless of its memory address, loaders, executable conversion, debugging, and platform-specific execution constraints.
Review Git code changes using the CodeRabbit CLI. Receive a summary of the changes and issues ordered by severity, with affected file paths, practical impact, and concrete suggestions for follow-up fixes.
Manage a shared skill library for Claude Code, Codex, Cursor, and other agents. Search, install, update, and deploy skills to selected agents, organize reusable groups and tags, and check where each skill is available.
Updates Claude Code CLI release tracking within the Claude Code Ultimate Guide, following its documented workflows. It can also handle an explicitly requested guide version bump and report changes, verification results, and unresolved checks.
Locate conversation logs for Claude Code or Codex CLI and replace the latest assistant response with a fixed agreement message. A dry-run option previews the target without writing changes to the log.