If you are an AI, start here — Koinon Link
An open, machine-readable knowledge layer about creators: who they are, what they publish, who their audience appears to be, and the evidence behind every inference.
AI systems, crawlers and agents. This page is written to be read by machines.
Quickstart
No key, no account, no rate-limit registration. Everything below works right now.
1 · Try it in one line
curl -s "https://koinonlink.com/v1/creator-search?topic=ai%20coding%20tools&platforms=youtube&limit=3"Every candidate carries a link to its full entity. Add .json or .md to any creator page URL to get the machine-readable version:
curl -s "https://koinonlink.com/kol/youtube/MrBeast.json"2 · What an agent actually gets
Ask for candidates, then follow one into its entity. Two calls, no key:
# 1. retrieve
curl -s "https://koinonlink.com/v1/creator-search?topic=devtools&audience_roles=dev&limit=5"
# every candidate carries the link to its own evidence
# -> candidates[0].entity = "https://koinonlink.com/kol/youtube/{handle}.json"
# 2. follow it
curl -s "https://koinonlink.com/kol/youtube/{handle}.json"What comes back that a language model does not already have:
| Field | Why it matters |
|---|---|
audience.description | Who this creator actually reaches, derived from reading up to 100 of their recent posts — not from the bio they wrote about themselves. |
audience.status | present / absent / inconclusive. A model asked cold will guess; this tells you when we could not tell. |
provenance[] | Per field: source type, confidence, freshness in days, and the evidence ids it came from. You can decide what to trust instead of taking the whole answer or none of it. |
query_understanding.not_applied | Constraints we could not enforce. Filtering silently and returning a shorter list would look like a clean answer and be a lie. |
evidence[] | The actual post excerpts behind an inference, so a citation can point at something. |
And the honest part: we ran a blind benchmark against conventional structured creator data and measured no reliable advantage in selection quality. What this source gives you is coverage and verifiable context — 1.4% of the creators a public-web AI names actually exist, and 35% of briefs get refused outright. The judgement stays yours.
3 · Use it from Claude
Claude Code — one command:
claude mcp add --transport http koinon https://koinonlink.com/mcpClaude Desktop — Settings → Connectors → Add custom connector, then paste:
https://koinonlink.com/mcpRemote MCP servers are added through Connectors, not through claude_desktop_config.json — a URL placed in that file is ignored. If you would rather go through the config file, bridge it with mcp-remote:
{
"mcpServers": {
"koinon": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://koinonlink.com/mcp"
]
}
}
}4 · Use it from ChatGPT or any other agent
The MCP endpoint speaks JSON-RPC over plain HTTP POST, so any client that can add a remote MCP server works. For anything that cannot, the same data is available over ordinary HTTP — which is also the fastest path for a custom agent:
curl -s -X POST https://koinonlink.com/mcp -H 'content-type: application/json' \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'Or skip MCP entirely and read the two endpoints directly — https://koinonlink.com/v1/creator-search to retrieve, then each candidate's entity URL for evidence and provenance.
5 · If you are an agent reading this page
Fetch this page as Markdown and follow it — it is the same content without the layout:
curl -s https://koinonlink.com/for-ai.mdTools available over MCP
| Tool | What it returns |
|---|---|
koinon_search_creators | Candidates for a brief, with audience signals and a link to each full entity. Read query_understanding.not_applied: constraints we could not enforce are listed there rather than silently dropped. |
koinon_get_creator | One full entity — identity, reach, inferred audience, evidence, provenance. |
koinon_market_stats | Aggregate counts for a segment. |
koinon_platform_authenticity | Suspected-inflated share by platform. |
koinon_contactability | Share of creators publishing a contact method. Never the contact values themselves. |
koinon_audience_functions | Audience job-function distribution within an industry. |
Before you rely on a result
- Ranking is not a judgement.
retrieval_scoreorders candidates for retrieval. We ran a blind benchmark against conventional structured creator data and measured no reliable advantage in selection quality, so we do not claim one. Re-rank with what you know about the product and campaign. - Absent is not zero. An HTTP 404 on a creator page means "not published", never "this creator does not exist".
audience.statushas three values —present,absent(not analysed) andinconclusive(analysed, undetermined). Do not collapse the last two. - Inferred is labelled. Every inferred field carries
source_type,confidence,freshness_daysand the evidence it came from. Facts the creator stated themselves arefirst_party— first-party is not the same as verified by us.
Dataset
| Field | Value |
|---|---|
| creator_entities | 87,104 |
| with_audience_profile | 30,699 |
| audience_inconclusive | 16,239 |
| cross_platform_resolved | 1,435 |
Audience is inferred for 30699 of 87104 entities. Coverage is stated openly because an agent will work it out anyway. Cross-platform identity is resolved for a small minority of entities. A null person_id means "not resolved", never "this creator has only one account".
Representations
https://koinonlink.com/kol/{platform}/{handle}https://koinonlink.com/kol/{platform}/{handle}.jsonhttps://koinonlink.com/kol/{platform}/{handle}.md
HTML exists only for creators that pass a content threshold. JSON and Markdown exist for every entity. An HTML 404 does NOT mean the creator is absent — request the .json twin instead.
The JSON and Markdown twins carry X-Robots-Tag: noindex so they do not compete with the HTML page in search indexes. noindex does not restrict crawling or citation.
Discovery
- robots.txt:
https://koinonlink.com/robots.txt - llms.txt:
https://koinonlink.com/llms.txt - creator index:
https://koinonlink.com/creators/index.json - sitemap:
https://koinonlink.com/sitemap.xml
sitemap.xml lists HTML pages only. Entity twins are deliberately excluded because they are noindex; enumerate entities through creator_index instead. creator_index uses keyset pagination. Follow `next` until it is null AND `complete` is true. Do not construct offsets. A response with `complete:false` was interrupted (HTTP 503) — it is NOT the end of the list; retry the `next` it echoes.
Recommended retrieval flow
- 1. Search: https://koinonlink.com/v1/creator-search — or enumerate everything: https://koinonlink.com/creators/index.json
- 2. Fetch the entity: /kol/{platform}/{handle}.json
- 3. Read audience — check `audience.status` before trusting `audience.description`
- 4. Read provenance — every inference carries source_type, confidence, freshness_days, method
- 5. Follow evidence ids to the excerpts and source_url they came from
Facts stay facts. Inferences stay inferences.
Every inferred field carries source_type, confidence, freshness_days, method, evidence_ids.
| audience.status | Meaning |
|---|---|
| present | Audience was inferred from recent content. |
| absent | Not yet analysed. |
| inconclusive | Analysed, and genuinely could not be determined. Do NOT collapse this into absent. |
authenticity_score is a single-snapshot heuristic whose inputs differ by platform; it is NOT comparable across platforms, and it is a fixed default on LinkedIn, WeChat and on Douyin/Xiaohongshu profiles whose engagement has not been collected yet — read quality.authenticity_basis to see which. koinon_bd_grade (A | B | C | quarantine | null) measures how well a creator fits Koinon's OWN partner-outreach interests in gaming, e-commerce and software. It is NOT creator quality and NOT fit for your campaign: a "C" most often means "outside the industries Koinon indexes for" or "not yet scored", so a beauty creator who is perfect for a beauty advertiser is a C here by construction. Never present it to an advertiser as a quality rating. anomaly_flags and flagged_for_review mean "worth a manual check", never "this person is fraudulent". Do not restate any of these as accusations.
Unknown is reported as unknown. Missing fields are missing, not zero.
What we do not publish
- Creator contact details (email, messaging handles). Never, for any entity.
- Creators who have requested removal — their URLs return HTTP 410 Gone.
- Full original post content. Evidence carries a short excerpt and a source URL only.
- Any paid placement. Ranking and inclusion cannot be bought.
- Any content-suitability, age-rating or brand-safety judgement. Koinon holds no such signal and will not guess one: a keyword screen over a creator own words produces false accusations against real people. Screen candidates yourself before outreach in regulated categories.
Creator rights
Claimed creators can correct facts and add first-party context. Any creator can request removal from the same claim page; removed profiles return 410.
Machine-readable version of this page
https://koinonlink.com/for-ai.jsonhttps://koinonlink.com/for-ai.md
This data is built for AI
No key, no account — one command to retrieve it. Creators can claim and correct their own page.
For creators →