Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Shared HomeRail rules for AI agents operating the local-source release candidate: service roles, environment variables, secrets, provider boundaries, Docker callback networking, update expectations, and safe command behavior. Use whenever installing, configuring, upgrading, troubleshooting, or running HomeRail through other HomeRail skills.
Skill 进化器（饕餮）— 通过"吞噬"并分析其他 skill 的优势来强化目标 skill。 当用户想要：整合两个 skill、用一个 skill 优化另一个、对比分析两个 skill 的优劣、 把某个 skill 的优点提炼到另一个 skill 中、或者说"把 X 喂给 Y"、"用 X 来优化 Y"、 "整合这两个 skill"、"吃掉这个 skill"、"skill 进化"、"skill 升级"、"合并 skill" 等意图时，必须触发此 skill。即使用户没有明确说"饕餮"，只要涉及到两个 skill 之间的 能力迁移、对比分析、或优势提取，都应该使用此 skill。
Add or change a built-in Synergy primary agent, subagent, hidden reviewer, prompt, visibility rule, delegation group, model role, or permission profile. Use for requests about built-in agent definitions under packages/synergy/src/agent; do not use for user-configured, plugin, or external agents.
Add or modify a first-party Synergy tool, its Zod parameters, execution behavior, capability taxonomy, exposure, permission boundary, attachments, or Web tool-card registration. Use for packages/synergy/src/tool and the corresponding packages/ui registrations; use plugin docs for plugin-owned tools.
Add, modify, or review Synergy Plugin API 3 definitions, generated manifests, plugin-kit builds, installation transactions, capability approval, process runtimes, operations/events/hooks, marketplace behavior, or trusted UI contribution lifecycle.
How to write system prompts and instructions for GPT-5.3-Codex. Use when constructing or tuning prompts targeting Codex 5.3.
Cheat sheet + workflow for launching interactive coding-agent CLIs (Claude Code, Gemini CLI, Codex CLI, Cursor CLI, and pi itself) via the interactive_shell overlay, headless dispatch, or monitor mode. Use for TUI agents and long-running processes that need supervision, fire-and-forget delegation, or event-driven background monitoring. Regular bash commands should use the bash tool instead.
Audit TSZ repo iteration speed and quality. Use when asked to deep-dive repo health, identify avoidable delays, improve agent workflows, reduce CI/PR churn, add guardrails, choose skills to install or create, or turn recurring TSZ process failures into focused tooling/docs PRs.
Use when designing, deploying, or debugging a Butterbase Agent (declarative LLM/tool graph), registering an MCP server for tool use, or wiring access controls and rate limits. Agents are first-class app resources defined by a `graph_spec` and invoked over `/v1/<app_id>/agents/<name>/runs`.