starraider/ai-skills-plugin
Portable Agent Skills for AI-harness diagnosis, AGENTS.md guidance, prompt design, IDE-native agents, hooks, slash commands, workflow creation, skill authoring, and Agent Plugin creation.
Create, package, migrate, release, or validate portable Agent Plugins 1.0.0 containing Agent Skills and optional MCP servers. Use when building an agent plugin, adding plugin.json or mcp.json, converting a client-specific plugin to the open standard, producing a clean archive, or auditing v1 conformance. Do not use for purely client-owned installation or enablement.
Diagnose how to improve an AI coding IDE or agent harness by interviewing the user, identifying required capabilities, selecting a prompt, AGENTS.md, Agent Skill, slash command, workflow, hook, specialized agent, Agent Plugin, or a justified combination, and producing a paste-ready implementation prompt that invokes this plugin's relevant builder skills. Use when the desired harness improvement or mechanism is unclear; do not use when the artifact is already chosen and the user wants direct implementation.
Use when the user wants to create a specialized agent for Antigravity, Codex, ChatGPT, Zed, OpenCode, Qoder, Orca, Cursor, GitHub Copilot / VS Code, Windsurf, Kiro, or Google Agents CLI. Require the target product and the agent's specific capabilities, then create the smallest supported agent definition and give product-appropriate verification steps. Do not treat skills, rules, or generic prompts as agents unless the selected product documents that model.
Use when the user wants to create, replace, or improve an AGENTS.md instruction file for Antigravity, Codex, ChatGPT, Zed, OpenCode, Qoder, or Orca. Establish the target product and scope, inspect the repository for evidence, create concise project-specific guidance, and give target-appropriate discovery checks. Do not use for skills, native rules, custom agents, or generic prompts unless an AGENTS.md is explicitly needed.
Use when deciding whether a lifecycle or event hook is the right solution, or when creating or revising one for Antigravity, Codex, ChatGPT, Zed, OpenCode, Qoder, Orca, Cursor, GitHub Copilot / VS Code, Windsurf, Kiro, or Google Agents CLI. Select exactly one target and documented event model; recommend a Skill, plugin/app, permission policy, task, automation, or CI when a hook is unsupported or unsuitable.
Create or improve a ready-to-use LLM prompt from a user's goal. Use when the user asks to write, optimize, structure, debug, evaluate, or template a prompt for a general-purpose or reasoning language model, including prompts for code, data, summaries, instructions, creative work, evaluations, or tool use. Do not use for Stitch-specific UI prompts; route those to an installed Stitch-focused Skill.
Use when creating, improving, validating, or packaging an Agent Skill. Covers standards-compliant SKILL.md authoring, bundled and standalone documentation, progressive disclosure, scripts, templates, and evaluation.
Use when deciding whether a reusable task should be a slash command and when creating or revising a target-native command or the documented alternative for Antigravity, Codex, ChatGPT, Zed, OpenCode, Qoder, Orca, Cursor, GitHub Copilot / VS Code, Windsurf, Kiro, or Google Agents CLI (Gemini CLI).
Use when deciding whether a repeatable workflow is appropriate and when creating or revising a target-native workflow for Antigravity, Codex, ChatGPT, Zed, OpenCode, Qoder, Orca, Cursor, GitHub Copilot / VS Code, Windsurf, Kiro, or Google Agents CLI.