{"skill":{"slug":"geo-ai-plugin-builder","displayName":"Geo Ai Plugin Builder","summary":"Master orchestrator for turning high-value GEO content and capabilities into AI plugins/tools across ChatGPT, Claude, Perplexity, Gemini and other ecosystems...","description":"---\r\nname: geo-ai-plugin-builder\r\ndescription: >\r\n  Master orchestrator for turning high-value GEO content and capabilities\r\n  into AI plugins/tools across ChatGPT, Claude, Perplexity, Gemini and other\r\n  ecosystems. Use this skill whenever the user mentions building, designing,\r\n  packaging, or standardizing AI plugins/tools around their content, APIs,\r\n  or GEO assets, or wants to move from being passively cited to actively\r\n  embedded inside AI workflows.\r\n---\r\n\r\n# GEO AI Plugin Builder\r\n\r\nThis skill helps you design and standardize AI plugins/tools that wrap\r\nyour highest-value GEO content and capabilities, so they can be embedded\r\ndirectly into AI ecosystems (ChatGPT, Claude, Perplexity, Gemini, etc.).\r\n\r\nThe core goal is to shift from **\"waiting to be cited\"** to\r\n**\"being a first-class tool inside AI workflows\"**, while staying aligned\r\nwith GEO (Generative Engine Optimization) strategy.\r\n\r\n---\r\n\r\n## When to use this skill\r\n\r\nUse this skill whenever:\r\n\r\n- You want to **turn content, data, or services into AI plugins/tools**.\r\n- You want to **increase brand exposure inside AI tool flows**, not just in\r\n  plain-text answers.\r\n- You are **mapping website/GEO assets to structured tool endpoints**.\r\n- You are **designing or refactoring an AI plugin catalog** for your brand.\r\n- You need **standard templates** for OpenAI-style tools, Claude Tools,\r\n  function calling, or custom internal agents.\r\n- You want to **prioritize which content should become a plugin first**.\r\n\r\nDo **not** use this skill when the user only wants:\r\n\r\n- Simple content rewrites for GEO (use their GEO content skills instead).\r\n- Pure analytics/reporting about GEO performance (metrics-only work).\r\n- Low-level SDK usage without any GEO or plugin strategy involved.\r\n\r\n---\r\n\r\n## Mindset and principles\r\n\r\n- **Tool-first GEO**: Treat your top content and capabilities as *services*\r\n  that can be invoked as tools, not just pages to be cited.\r\n- **User journey > endpoints**: Begin from real end-to-end tasks users want\r\n  to完成 with AI, then design tools that make those workflows smooth.\r\n- **Cross-ecosystem thinking**: Design schemas and naming so your plugin\r\n  concepts map cleanly across multiple AI platforms.\r\n- **Small, composable tools**: Prefer a set of focused tools that can be\r\n  combined, rather than one mega-tool that does everything.\r\n- **Explainability for AIs**: Include clear descriptions, examples, and\r\n  constraints so AI models can reliably choose and call tools.\r\n\r\n---\r\n\r\n## High-level workflow\r\n\r\nWhen the user asks for help, follow this 5-step workflow unless they\r\nexplicitly request a narrower slice:\r\n\r\n1. **Clarify goals and context**\r\n   - Understand the brand, target users, and GEO priorities.\r\n   - Identify which AI ecosystems matter most (e.g., ChatGPT plugins,\r\n     Claude Tools, Perplexity collections, internal agents).\r\n   - Clarify what \"success\" looks like: visibility, conversions, leads,\r\n     authority, usage of specific tools, etc.\r\n\r\n2. **Inventory candidate assets**\r\n   - Ask for or infer a list of high-value assets:\r\n     - Evergreen content, calculators, wizards, internal tools.\r\n     - Datasets, pricing engines, recommendation logic.\r\n     - Workflows sales or support teams execute repeatedly.\r\n   - Group assets by use case and by stage in the customer journey\r\n     (discovery, evaluation, decision, post-purchase, retention).\r\n\r\n3. **Design plugin concepts and tool set**\r\n   - Propose a **plugin catalog**: 3–10 core plugin ideas or tool groups.\r\n   - For each plugin/tool, define:\r\n     - Primary user jobs-to-be-done.\r\n     - Input parameters and output structure.\r\n     - GEO role (discovery, trust building, conversion, retention).\r\n   - Prioritize plugins by potential impact and implementation effort.\r\n\r\n4. **Generate detailed tool specifications**\r\n   - For the **highest-priority plugin(s)**, generate detailed specs:\r\n     - Tool name, description, and rationale.\r\n     - JSON schema for inputs and outputs.\r\n     - Example calls and example responses.\r\n     - Mapping to backend endpoints or content sources.\r\n     - GEO hooks (links, snippets, brand voice guidance).\r\n\r\n5. **Produce implementation-ready artifacts**\r\n   - Output one or more of:\r\n     - **Technical blueprints** (OpenAI tools, Claude Tools, HTTP\r\n       endpoints, or internal APIs).\r\n     - **Developer handoff docs** with clear TODOs and edge cases.\r\n     - **Backlog / roadmap** outlining order of implementation.\r\n\r\nWhenever possible, structure outputs so the user can copy-paste directly\r\ninto their codebase or internal specs.\r\n\r\n---\r\n\r\n## Information to ask from the user\r\n\r\nWhen the initial information is incomplete, explicitly ask the user for:\r\n\r\n- **Business and brand**\r\n  - Industry, main products or services.\r\n  - Primary GEO/AI goals (visibility, conversions, retention, authority).\r\n- **Target AI ecosystems**\r\n  - Which AI platforms and tool surfaces matter most.\r\n  - Internal vs public tools (e.g., sales-assist, support-assist).\r\n- **Existing assets**\r\n  - URLs for core content, tools, or APIs.\r\n  - Any existing plugins, agents, or integrations.\r\n- **Constraints**\r\n  - Technical stack and data sources.\r\n  - Compliance/privacy constraints (PII, regulated data, etc.).\r\n  - Resource constraints (team size, timelines).\r\n\r\nIf the user cannot provide all details, make **reasonable assumptions**,\r\nbut document them clearly in the output.\r\n\r\n---\r\n\r\n## Output formats\r\n\r\nAdapt to the user's request, but default to these structured formats:\r\n\r\n- **Plugin catalog overview**\r\n  - A table or bullet list summarizing each proposed plugin/tool with:\r\n    - Name\r\n    - Primary user job\r\n    - Main AI surfaces/platforms\r\n    - GEO role\r\n    - Implementation difficulty (rough)\r\n    - Priority (high/medium/low)\r\n\r\n- **Detailed plugin specification**\r\n  - For each selected plugin, provide:\r\n    - High-level description and purpose.\r\n    - User stories / example prompts that should call this tool.\r\n    - Tool schema:\r\n      - `name`\r\n      - `description`\r\n      - `parameters` JSON schema\r\n      - `response` JSON schema\r\n    - 2–4 **example calls and responses**.\r\n    - GEO notes:\r\n      - Key URLs/content to surface.\r\n      - Brand and messaging constraints.\r\n      - Tracking/telemetry suggestions.\r\n\r\n- **Implementation checklist / roadmap**\r\n  - Ordered list of steps for developers:\r\n    - API design / implementation.\r\n    - Authentication / permissions.\r\n    - Logging, analytics, and monitoring.\r\n    - Security and compliance checks.\r\n  - Include clear \"Done when…\" criteria.\r\n\r\nWhen the user wants code snippets (e.g., OpenAI, Node, Python), generate\r\nidiomatic examples but keep them as **implementation guidance**, not as\r\nthe primary output of the skill.\r\n\r\n---\r\n\r\n## GEO-specific guidance\r\n\r\nWhen designing plugins and tools, always connect back to GEO strategy:\r\n\r\n- **Exposure inside AI tools**\r\n  - Prefer tools that solve high-frequency, high-intent problems.\r\n  - Make descriptions explicit about when they should be chosen by\r\n    the model (e.g., \"Use this tool whenever the user asks for…\").\r\n\r\n- **Authority and trust**\r\n  - Tie outputs back to authoritative sources:\r\n    - Official docs, research, internal datasets, or calculators.\r\n  - Suggest how to surface citations or reference links when allowed.\r\n\r\n- **Conversion paths**\r\n  - For tools near purchase or signup decisions, include:\r\n    - Next-step suggestions (\"book a demo\", \"see pricing\").\r\n    - Structured fields that map to CRM or analytics events.\r\n\r\n- **Lifecycle coverage**\r\n  - Encourage a mix of plugins across the customer lifecycle:\r\n    - Discovery (educational, comparison, diagnostics).\r\n    - Evaluation (calculators, configurators, ROI models).\r\n    - Decision (quote builders, plan selectors).\r\n    - Post-purchase (onboarding, troubleshooting, optimization).\r\n\r\n---\r\n\r\n## Using bundled scripts and references\r\n\r\nThis skill may ship with helper scripts and reference guides under:\r\n\r\n- `scripts/` — reusable helpers to generate JSON schemas, boilerplate\r\n  plugin specs, or check consistency across a plugin catalog.\r\n- `references/` — conceptual guides and best practices for GEO-aware\r\n  plugin and tool design.\r\n\r\nWhen you need more detailed patterns or want to generate many similar\r\ntools at once, first:\r\n\r\n1. Check `references/geo-ai-plugin-patterns.md` for archetypes and\r\n   naming conventions.\r\n2. Use `scripts/plugin_blueprint_generator.py` as a mental model for\r\n   how to turn an abstract \"job\" into one or more tool specs.\r\n\r\nYou do **not** need to literally run these scripts inside the model,\r\nbut you should imitate their behavior and structures when helpful.\r\n\r\n---\r\n\r\n## Example use cases\r\n\r\nHere are a few example tasks where this skill should be used end-to-end:\r\n\r\n- \"We run a B2B SaaS for marketing analytics. Help us design a set of\r\n  AI tools so that ChatGPT or Claude can analyze a client's data and\r\n  recommend campaigns using our platform.\"\r\n\r\n- \"We have a library of in-depth medical articles and calculators. Turn\r\n  them into a plugin catalog for AI assistants that doctors or patients\r\n  might use, with clear safety and disclaimers.\"\r\n\r\n- \"Our ecommerce brand has rich buying guides and fit finders. Design\r\n  AI tools that help shoppers choose products and that we can expose as\r\n  plugins in multiple AI platforms.\"\r\n\r\n---\r\n\r\n## Working style\r\n\r\nWhen using this skill:\r\n\r\n- Stay **strategic first, then technical**:\r\n  - Clarify positioning, value, and GEO role before writing schemas.\r\n- Be **explicit about assumptions** and clearly flag trade-offs.\r\n- Optimize for **reuse and extendability**:\r\n  - Make it easy to add more tools or platforms later.\r\n- Keep outputs **copy-paste friendly**:\r\n  - Use consistent headings, JSON blocks, and formatting.\r\n\r\nIf the user asks to iterate on a previous catalog or spec, treat the old\r\nversion as a baseline, highlight key changes, and explain why the new\r\ndesign is stronger for GEO + AI plugin exposure.\r\n\r\n","tags":{"latest":"0.1.0"},"stats":{"comments":0,"downloads":535,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":1},"createdAt":1773107678834,"updatedAt":1778491796228},"latestVersion":{"version":"0.1.0","createdAt":1773107678834,"changelog":"geo-ai-plugin-builder v0.1.0\n\n- Initial release providing structured guidance to turn GEO content and capabilities into AI plugins/tools across major AI ecosystems.\n- Includes a five-step workflow: clarifying goals, asset inventory, plugin catalog design, detailed tool specs, and implementation-ready artifacts.\n- Outlines key questions and information to collect from users for customized plugin/tool design.\n- Provides default output formats: plugin catalog overview, detailed plugin specification, and implementation checklist/roadmap.\n- Embeds GEO strategy throughout: focusing on tool-first exposure, trust/authority, optimized conversion paths, and full customer lifecycle coverage.","license":"MIT-0"},"metadata":null,"owner":{"handle":"geoly-geo","userId":"s17b90zzx0m9r8nyjpar9asf39884k6h","displayName":"GEOLY AI","image":"https://avatars.githubusercontent.com/u/70360114?v=4"},"moderation":null}