study8677

Perform a release-readiness review by locating the previous release tag from remote tags and auditing the diff (e.g., v1.2.3...<commit>) for breaking changes, regressions, improvement opportunities, and risks before releasing openai-agents-python.

Analyze main branch implementation and configuration to find missing, incorrect, or outdated documentation in docs/. Use when asked to audit doc coverage, sync docs with code, or propose doc updates/structure changes. Only update English docs under docs/** and never touch translated docs under docs/ja, docs/ko, or docs/zh. Provide a report and ask for approval before editing docs.

Improve test coverage in the OpenAI Agents Python repository: run `make coverage`, inspect coverage artifacts, identify low-coverage files, propose high-impact tests, and confirm with the user before writing tests.

Run python examples in auto mode with logging, rerun helpers, and background control.

Create a PR title and draft description after substantive code changes are finished. Trigger when wrapping up a moderate-or-larger change (runtime code, tests, build config, docs with behavior impact) and you need the PR-ready summary block with change summary plus PR draft text.

Run the mandatory verification stack when changes affect runtime code, tests, or build/test behavior in the OpenAI Agents Python repository.

Use when working with the OpenAI API (Responses API) or OpenAI platform features (tools, streaming, Realtime API, auth, models, rate limits, MCP) and you need authoritative, up-to-date documentation (schemas, examples, limits, edge cases). Prefer the OpenAI Developer Documentation MCP server tools when available; otherwise guide the user to enable `openaiDeveloperDocs`.

Bootstraps a new multi-agent repository from the Antigravity template via `init_agent_repo`. Supports quick scaffold and full runtime profile setup including MCP toggle, swarm preference, sandbox type, and optional git init. LLM configuration is handled later by ag-setup.

One-click initialization of a multi-agent repository from the Antigravity template. Use this skill when users want to scaffold a new project quickly (`quick` mode) or with runtime defaults (`full` mode) including MCP toggle, swarm preference context, sandbox type, and optional git init. LLM configuration is handled later by ag-setup.

Exposes graph-based retrieval as a tool capability via `query_graph`. Reads normalized graph store files, builds a query-relevant subgraph, and returns LLM-friendly semantic triples with replayable evidence metadata.

High-level deployment wrapper over Antigravity core with graph-first knowledge injection and all-file support. Exposes `refresh_filesystem` and `ask_filesystem` for building and querying the knowledge graph.

Performs deep research on a topic via `deep_research`. Simulates a multi-step research process and returns a comprehensive research result as a string.

生成一份对外可分享、脱敏的 AI-Native 开发者 README。 量化展示我对 Claude Code + Codex CLI 的使用深度、AI 协作风格、项目与领域分布、 兴趣主题，以及与 GitHub 提交的产出关联。 Trigger when the user says: "生成我的 AI 档案" / "做一份 AI-native README" / "分析我的 Claude 使用情况" / "总结我的 AI 使用" / "build my AI usage profile" / "summarize my Claude / Codex history" / "生成开发者画像". 全程本地、只读、默认匿名、不上传任何数据。

Use AVC (Agent View Controller) to present complex execution plans, architecture changes, and multi-step operations as interactive visual UIs. Instead of dumping walls of text, pipe structured JSON to `avc` for human visual review and confirmation. The human can drag-to-reorder, edit, skip, delete, and add steps before confirming.

引导式「架构共创」教练。当用户面对一个新项目 / 新系统、想在动手写代码前把架构想清楚时使用, 也适用于系统设计面试练习。它不直接给方案,而是通过分阶段的深度提问 (一句话定位 → 业务本质与范围 → 灵魂六问 → 信封背面估算 → 质量属性取舍 → 关键决策追问) 引导用户一步步收敛出:架构全景图、数据模型、ADR 决策记录、规模化瓶颈与演进路线。 方法论与案例知识源自开源仓库 awesome-architecture。 触发词:设计架构、系统设计、技术方案、架构评审、"我想做一个…该怎么设计"、system design。

Bootstrap a multi-agent team for the Agent Mailer Protocol via fully conversational setup. In claude code or codex chat, asks the user what agents they want, drafts system_prompts, confirms, then writes team.yaml and runs `agent-mailer up-team` to register agents and generate launcher scripts. Idempotent — re-running detects an existing team.yaml and only refreshes what's needed. The skill never launches terminals; that's start-team.sh's job. Use when the user invokes /zudui or 组队 / 开工 to bring up a team. Pairs with shangban (上班) and xiaban (下班) per agent.