{
    "specVersion": "1.0",
    "host": {
        "displayName": "OrganiKPI",
        "identifier": "did:web:organikpi.com"
    },
    "entries": [
        {
            "identifier": "urn:air:organikpi.com:server:mcp",
            "displayName": "OrganiKPI MCP Server",
            "type": "application/mcp-server-card+json",
            "url": "https://organikpi.com/.well-known/mcp/server-card.json",
            "description": "Read-only MCP server: search the blog, read any page as markdown, list the open AI search research datasets.",
            "capabilities": [
                "search_content",
                "fetch_markdown",
                "list_datasets"
            ],
            "representativeQueries": [
                "find research on which sources AI engines cite",
                "read an article about generative engine optimization as markdown",
                "list open datasets about AI search citations"
            ]
        },
        {
            "identifier": "urn:air:organikpi.com:skill:read-content",
            "displayName": "Read any page as markdown",
            "type": "application/ai-skill+md",
            "url": "https://organikpi.com/.well-known/agent-skills/read-content.md",
            "description": "How to fetch the full markdown of any page (saves tokens vs HTML)."
        },
        {
            "identifier": "urn:air:organikpi.com:skill:submit-contact",
            "displayName": "Submit a contact form on the user's behalf",
            "type": "application/ai-skill+md",
            "url": "https://organikpi.com/.well-known/agent-skills/submit-contact.md",
            "description": "How to send a contact request via the public REST endpoint."
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:ai-citation-accuracy-study",
            "displayName": "AI Citation Accuracy Dataset",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/10/organikpi-ai-citation-accuracy-2026-10.csv",
            "description": "2,690 citation-claim pairs from ChatGPT, Gemini, and Copilot, each judged as SUPPORTED, PARTIAL, NOT_SUPPORTED, or UNREACHABLE against the cited page text. Covers 160 product-comparison queries collected October 2026.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/ai-citation-accuracy-study/",
                "json": "https://organikpi.com/wp-content/uploads/2026/10/organikpi-ai-citation-accuracy-2026-10.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/ai-citation-accuracy"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:self-promotional-listicles-ai-search",
            "displayName": "Vendor self-ranking lists in AI answers",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-vendor-self-ranking-lists-2026-09.csv",
            "description": "Every source ChatGPT, Microsoft Copilot, Google Gemini and Google AI Mode cited across 160 best X buyer prompts, three runs each (September 2026), typed by publisher, with where vendor-written lists rank their own product and whether the answer recommended it.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/self-promotional-listicles-ai-search/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-vendor-self-ranking-lists-2026-09.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/vendor-self-ranking-lists"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:chatgpt-query-fanout-study",
            "displayName": "ChatGPT query fan-out, August 2026",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/fanout_queries.csv",
            "description": "219 search queries ChatGPT wrote for itself across 121 answers, with Google Keyword Planner status, length, and the companies, products and publications named in each query but absent from the user's prompt.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/chatgpt-query-fanout-study/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/fanout_queries.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/chatgpt-query-fanout"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:ai-overview-citations-best-x-queries",
            "displayName": "AI Overview citations on 160 best-X buyer queries, September 2026",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-aio-best-x-citations-2026-09.csv",
            "description": "Every citation Google's AI Overview returned for 160 'best X' commercial queries (US, desktop, September 2026): 1,208 cited URLs with domain, our source classification, whether the cited page is itself a ranked list, and whether the domain also ranks in the query's organic top 10.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/ai-overview-citations-best-x-queries/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-aio-best-x-citations-2026-09.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/ai-overview-best-x-citations"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:which-linkedin-urls-ai-engines-cite",
            "displayName": "Five-engine AI citations on 100 professional prompts, August 2026",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-linkedin-ai-citations-2026-08.csv",
            "description": "Every URL returned by five AI engines (Google AI Mode, ChatGPT, Copilot, Gemini, Perplexity) for 100 professional-intent prompts, US, August 2026: 4,008 URL rows across 456 valid responses, with the citation field, domain, LinkedIn URL type, and utm_source tag per row.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/which-linkedin-urls-ai-engines-cite/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-linkedin-ai-citations-2026-08.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/linkedin-ai-citations"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:chatgpt-reddit-citation-collapse",
            "displayName": "Five-engine AI citations on 100 consumer prompts plus ChatGPT A/B, August 2026",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-reddit-chatgpt-citations-2026-08.csv",
            "description": "Every URL returned by five AI engines for 100 consumer prompts (US, August 2026), plus a 50-record paired A/B where ChatGPT prompts appended 'What do real users say about it?': 10,118 URL rows across 550 responses, with the citation field and domain per row.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/chatgpt-reddit-citation-collapse/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-reddit-chatgpt-citations-2026-08.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/reddit-chatgpt-citations"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:chatgpt-citation-variance-study",
            "displayName": "ChatGPT citation variance across 4 same-day runs, August 2026",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-chatgpt-citation-variance-2026-08.csv",
            "description": "Every source ChatGPT consulted across a 58-prompt main run and three same-day replicate runs on a 28-prompt subset (US, gpt-5-6, August 2026): 2,642 source rows with the attributed/dropped flag, web-search trigger, and ChatGPT's own fan-out subqueries per response.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/chatgpt-citation-variance-study/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-chatgpt-citation-variance-2026-08.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/chatgpt-citation-variance"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:ai-engines-different-recommendations",
            "displayName": "AI engine brand recommendations on best X buyer prompts",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-ai-engine-recommendations-2026-09.csv",
            "description": "Every brand ChatGPT, Microsoft Copilot, Google Gemini and Google AI Mode recommended across 160 best X buyer prompts, three runs each (September 2026), with cited domains and shopping modules.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/ai-engines-different-recommendations/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-ai-engine-recommendations-2026-09.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/ai-engine-recommendations"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:ads-in-google-ai-mode",
            "displayName": "Google AI Mode ads and organic squeeze query set",
            "type": "text/csv",
            "url": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-ai-mode-squeeze-2026-07.csv",
            "description": "160 commercial queries across 8 verticals (Google US desktop, July 2026): AI Overview presence, domains cited in each AI Overview, ad counts, and how far the first organic result was pushed down the page.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/ads-in-google-ai-mode/",
                "json": "https://organikpi.com/wp-content/uploads/2026/09/organikpi-ai-mode-squeeze-2026-07.json",
                "license": "CC BY 4.0",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/ai-mode-organic-squeeze"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:ai-mode-text-fragments-dead-153425-citations",
            "displayName": "AI Mode text-fragment citation corpus (153,425 citations)",
            "type": "text/html",
            "url": "https://github.com/danishashko/ai-citation-patterns",
            "description": "153,425 Google AI Mode citations collected by OrganikPI before and after Google removed text fragments, used to measure how citation linking behavior changed.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/seo-strategy/ai-mode-text-fragments-dead-153425-citations/",
                "json": "https://github.com/danishashko/ai-citation-patterns",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/ai-citation-patterns"
            }
        },
        {
            "identifier": "urn:air:organikpi.com:dataset:decoded-42971-ai-citations-google-research",
            "displayName": "Google AI Mode citation-sentence corpus (42,971 citations)",
            "type": "text/html",
            "url": "https://github.com/danishashko/grounding-citation-analysis",
            "description": "42,971 AI Mode citations collected and decoded by OrganikPI to analyze how Google selects which sentences to cite, including position, structure, and phrasing patterns of cited passages.",
            "tags": [
                "dataset",
                "ai-search"
            ],
            "metadata": {
                "article": "https://organikpi.com/blog/geo-ai-search/decoded-42971-ai-citations-google-research/",
                "json": "https://github.com/danishashko/grounding-citation-analysis",
                "repo": "https://github.com/danishashko/ai-search-datasets/tree/main/grounding-citation-analysis"
            }
        }
    ]
}