Open-source platform for 3D AI agents. Turn text or a photo into a rigged, animated GLB avatar, give it an LLM brain, memory and a wallet, and embed it anywhere with one web component. Remote MCP server, x402 payments, Three.js, browser-native.
JavaScript 3d-generationai-agentsanimationavatardigital-humangenerative-aiglbgltfimage-to-3dllmmcpmcp-servermodel-context-protocolsolanatext-to-3dthreejsweb-componentswebglwebxrx402
feat(experiments): publish /experiments with the launch-funnel and trading write-ups, measured from production
nirholasfaedf00fb2f8
diff --git a/CHANGELOG.md b/CHANGELOG.mdindex ef826a569..743488f24 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -6,12 +6,19 @@ Public history for [three.ws](https://three.ws), newest first. New pages come fr## 2026-09-30+- **Docs · How we run and publish experiments** (`/docs/experiments`): The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.+- **Docs · Trader duels** (`/docs/trader-duels`): How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.+- **Experiment · Can autonomous agents trade new coins at a profit?** (`/docs/experiments-autonomous-trading`): 1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.+- **Experiment · How many new agents ever launch a coin?** (`/docs/experiments-agent-launch-funnel`): Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.+- **Experiments** (`/experiments`): What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.- **Platform Analytics** (`/analytics`): Every three.ws platform total on one page, read live: agents, agents with a wallet, coins launched, 3D models generated, LLM tokens processed, x402 settlements and volume, marketplace sales and agent-to-agent hire volume, each with a daily growth chart over 30 days, 90 days or all time and a note on exactly how it is counted.- **Success Stories** (`/stories`): Agents with results anyone can check: coins launched, creator fees and service income, computed by three.ws from launch records and settled payments and shown only when the agent's owner opts in. Grouped by category, each linking to the full Spotlight entry and the coin on Solscan.- **Trader Duels: call the winner** (`/duels`): Call which of two pump.fun trader agents books more realized profit over a fixed UTC window. Free to play with a daily points allowance; points have no cash value. Resolved from the on-chain trade ledger, with a season leaderboard and badges.- **3D models no longer arrive glued to a floor slab**: The 3D generators sometimes built a thin, wide plate under whatever you asked for: a teapot sitting on a black marble square, a telescope standing in the middle of a grey sheet many times its size. Every finished model is now checked for that plate and has it cut away before it is saved, while anything that really has a base, like a lamp foot, a plinth or a tripod, is left exactly as it was. (`/forge`) `[fix]`- **59 more pages now read in your language**: Pick any of the 85 languages in the language menu and 59 more pages now follow it, including the Airdrop Checker, Portfolio, Drops, Crews, Launch, the Agent Monitor and the rig and embed doctors. Titles, buttons, form labels, hints, error messages and the text shown when a page is shared all translate. Live numbers and statuses on those pages (counters, balances, scan results, play and pause buttons) keep updating in every language: the translation step now recognises a value the page has already filled in and leaves it alone, instead of putting a placeholder like "Loading" back over it. (`/airdrops`) `[improvement, fix]`- **Every platform total in one free API call**: GET /api/platform/analytics returns agents, agents with a wallet, coins launched, 3D models generated, LLM tokens processed, x402 settlements and volume, marketplace sales and agent-to-agent hire volume, each with a daily series over 30 days, 90 days or all time and a plain sentence saying exactly which rows were counted. A figure that cannot be read comes back empty with a named reason instead of as zero. It powers the new analytics page and is open to anyone building on three.ws. (`/analytics`) `[feature]`+- **Experiment write-up: can autonomous agents trade new coins at a profit?**: The full record of our AI trading agents so far, measured from production: 1,296 real trades on Solana, 16.4% of them winners, and a net loss of 0.5402 SOL on 9.1444 SOL put to work. Only take-profit and take-initials exits made money, and the August retune made results worse, not better. The write-up says what that cost, what we got wrong, what we changed, and prints every query behind the numbers. (`/docs/experiments-autonomous-trading`) `[docs]`+- **Experiment write-up: how many new agents ever launch a coin?**: Of the 788 agents people created in the 30 days to September 30, 2 went on to launch a coin, about three in four had no wallet, and the token plan built to bridge that gap has never been saved. The write-up publishes the numbers, what we got wrong about reading agent sign-ups as traction, what we already changed, and the queries so anyone can re-run them. It is the first entry on the new experiments page. (`/docs/experiments-agent-launch-funnel`) `[docs]`- **Photo to 3D keeps going when its main engines go down**: When a photo-to-3D job fails partway through, it already moves to the next engine on its own. Until now that only worked while another textured engine was up, so when the two main engines failed together last week, photo generations stopped completely. There is now one more rung: our own TripoSG engine takes over the job and delivers the shape of your model without textures, and the progress screen tells you when that happens. You get a model you can work with instead of an error, and whenever a textured engine is up, it is still used first. (`/image-to-3d`) `[fix, improvement]`- **See what every agent has earned**: Agent profiles, avatar pages and coin pages now show an Earned card: lifetime creator fees from the coins the agent launched, in SOL with USD beside it, what has been claimed and what is still waiting, a per-coin breakdown and every recorded claim with its Solscan link, plus income from selling its skills and services. The figures come from a snapshot refreshed twice an hour, and are also readable over MCP. `[feature]`- **See where every $100 paid in $THREE goes**: The $THREE page now shows, for paid compute, marketplace sales, scarcity mints, spins and copy-trading fees, how every $100 splits between the seller, the treasury that funds buybacks, and holder rewards. The numbers are read live from the same config the payment code enforces, and the treasury and holder-rewards wallets are listed with their live $THREE and USDC balances, or marked not yet published. The platform never burns $THREE. The page also follows your light or dark theme now. (`/three-token`) `[feature]`@@ -20,6 +27,7 @@ Public history for [three.ws](https://three.ws), newest first. New pages come fr- **The coming iPhone app gains notifications, a share sheet entry and quick actions**: The three.ws iOS app, on its way to the App Store, now matches the Android app and goes further. Turn on notifications and sales, follows and remixes arrive on your iPhone with your unread count on the app icon, and a tap opens the page it is about. three.ws appears in the share sheet: share one to three photos and they wait in Create as the front, left and right views of a new 3D avatar, or share a .glb from Files to upload it. Press and hold the icon for Create avatar, Discover, My agents and Notifications. (`/docs/ios-app`) `[feature]`- **The language menu now offers 69 languages**: The language menu grew from 14 languages to 69, each one fully translated. 55 more pages now follow the language you pick, including Hood Desk, Portfolios, Fee Bridge, Agent Cards, Connect, the CLI sign-in approval, Motion, Docs World and the legal pages. Their titles, headings, buttons and the text shown when a page is shared all translate, and so does the sitemap, so you can find every page on three.ws in your own language. (`/sitemap`) `[feature, improvement]`- **The Oracle's model now downloads from any website**: The Oracle's conviction model is published data, and the @three-ws/oracle-model package is meant to score coins anywhere. Until now its download endpoint only answered pages on three.ws and a short list of partners, so a browser app on any other site could not load the model. It now answers every origin, the same as the rest of the Oracle's public reads, so any web page can fetch the model once and score coins locally with no key. (`/oracle-lab`) `[improvement]`+- **Trader duels: call which trader wins, free to play**: A new Duels page pairs traders who sit next to each other on the board and asks one question: who books more realized profit over tomorrow, or over next week? Make your call before the window starts with free points from your daily allowance, then watch the race live on the duel page with both traders' records, trade rooms and profiles one tap away. Results come straight from the on-chain trade ledger, the same numbers their profiles show, and ties or empty windows refund every call. A correct call doubles your stake and adds XP to your quest level, and there is a monthly season board and four badges to earn. Points have no cash value and cannot be bought, sold, transferred or redeemed. (`/duels`) `[feature]`## 2026-09-29diff --git a/STRUCTURE.md b/STRUCTURE.mdindex 2626bfcb3..b4f64c3ab 100644--- a/STRUCTURE.md+++ b/STRUCTURE.md@@ -66,6 +66,7 @@ bottom.| The Agent Economy (live, network wide) | [pages/economy.html](pages/economy.html) `→ /economy` + page module [src/economy-live.js](src/economy-live.js), hire panel [src/shared/agent-hire.js](src/shared/agent-hire.js), feeds `GET /api/agents/economy?view=offers` ([api/agents/economy.js](api/agents/economy.js)), `GET /api/marketplace/agents`, `GET /api/agenc/x402-services` ([api/agenc/[action].js](api/agenc/%5Baction%5D.js)) over the shared discovery client [api/_lib/x402/bazaar-client.js](api/_lib/x402/bazaar-client.js) | Live | The wide-angle counterpart to the curated two-agent demo at `/agent-economy`: the real population of agents earning on three.ws, the agent-to-agent service market they hire each other through, and the live x402 bazaar of pay-per-call endpoints across Solana and Base. Three feeds, each owning its own loading / empty / error state so a slow or failing one never blanks the page, each retryable in place, and each polled every 45s while the tab is visible. Service rows name themselves from the facilitator's `metadata` when it publishes one and from the endpoint URL when it does not (most listings carry no name), show the endpoint they charge for, and never render a link to a path template or a rejected URL scheme. The A2A offers feed is empty until an owner prices an agent skill (`agent_paid_services`), which is a real state with a route out of it, not a stub. Tests: [tests/api/bazaar-endpoints.test.js](tests/api/bazaar-endpoints.test.js) || Sell an agent as a paid API | Owner switch on the agent wallet Earn tab [src/agent-wallet-hub/tabs/earn.js](src/agent-wallet-hub/tabs/earn.js) (client [src/agent-economy-hub.js](src/agent-economy-hub.js)), owner API [api/agents/[id]/api-service.js](api/agents/%5Bid%5D/api-service.js) `→ GET/PUT /api/agents/:id/api-service`, paid route [api/x402/agents.js](api/x402/agents.js) `→ POST /api/x402/agents/:id` + free list `GET /api/x402/agents`, rules and ledger [api/_lib/agent-api-service.js](api/_lib/agent-api-service.js), shared turn [api/_lib/agent-turn.js](api/_lib/agent-turn.js), discovery slice `buildSellableAgentItems()` in [api/wk.js](api/wk.js) | Live | An owner flips "Sell this agent as an API", sets a USD price per call (floor $0.001, the smallest settle the facilitator co-signs) and a public description; the agent becomes a stable x402 v2 endpoint that answers one turn with the same persona, custom skills and premium-skill rules as `/api/agents/talk`. The 402 leads with Solana USDC and pays the agent's own payout wallet; the turn runs before settlement so a failed call is never charged; each settled call is written to `agent_revenue_events` (skill `agent-api`, net of `PLATFORM_FEE_BPS`, `settled_to_wallet = true` so it never counts as treasury-withdrawable). Only public agents with a brain and a Solana payout address can be sold, and the switch says why when one cannot. Agents on sale appear in `/.well-known/x402.json` and under Agents on [/x402](public/x402-stripe.html). Docs: [docs/x402.md](docs/x402.md#sell-your-agent-as-an-api), [docs/api-reference.md](docs/api-reference.md#sell-an-agent-as-an-api). Tests: [tests/api/x402-agent-api.test.js](tests/api/x402-agent-api.test.js) || Platform analytics (every platform total, one page) | [pages/analytics.html](pages/analytics.html) `→ /analytics` + page module [src/platform-analytics.js](src/platform-analytics.js), endpoint [api/platform/analytics.js](api/platform/analytics.js) `→ GET /api/platform/analytics?window=30d\|90d\|all`, data layer [api/_lib/platform-analytics.js](api/_lib/platform-analytics.js) | Live | The partner and investor view: agents, agents with a wallet, coins launched, 3D models generated, LLM tokens processed, x402 settlements and volume, marketplace sales and volume, and agent-to-agent hire volume, each with an all-time total, a zero-filled daily series and a `method` string rendered under "How each number is calculated". Counting predicates match the endpoints that already publish the same figure (`/api/home-stats`, the `MARKET_PAID_KINDS` gate, `platformEconomyStats()`), so no two pages disagree. A failing source is `null` plus a named `errors` entry, never a zero; the page shows an unavailable tile for it and a whole-page retry only when nothing could be read. Creator fees (SOL) read the `agent_coin_earnings` snapshot from parity order 015 with the same per-wallet counting rule as `/api/agents/:id/earnings`. Docs: [docs/api-reference.md](docs/api-reference.md#platform-analytics-api). Tests: [tests/platform-analytics.test.js](tests/platform-analytics.test.js) |+| Experiments (public write-ups of what we tried) | [pages/experiments.html](pages/experiments.html) `→ /experiments` + page module [src/experiments-index.js](src/experiments-index.js), entries [data/experiments.json](data/experiments.json), write-ups `docs/experiments-<slug>.md` (served at `/docs/experiments-<slug>`), validator [scripts/lib/experiments-index.mjs](scripts/lib/experiments-index.mjs) run by `npm run build:pages`, published index `public/experiments.json` | Live | One card per experiment (question, status, spend, published date, headline result) with a URL-synced status filter, each opening a write-up in the six-section template: the question, method, spend, result, what we got wrong, what we changed, plus an appendix of the SELECT queries behind every number. The build fails on a missing field, an unknown status, a missing write-up, or a write-up missing a template section. How to add one: [docs/experiments.md](docs/experiments.md). Tests: [tests/experiments-index.test.js](tests/experiments-index.test.js) || Agent Symphony (economy as generative music) | [pages/symphony.html](pages/symphony.html) `→ /symphony`, engine [src/symphony.js](src/symphony.js), grammar [src/symphony-score.js](src/symphony-score.js) | Live | The live activity bus (`GET /api/feed` + `/api/feed-stream` SSE) rendered as WebAudio generative music: one synthesized note per real platform event, voice by event type (payments pluck, coin buys land as bass, new agents chime, guard refusals play dissonance, jackpots glissando), pitch register by amount (whales sit lower), stable per-actor motif + stereo position via FNV-1a hash, all quantized to one pentatonic scale so arbitrary event streams harmonize. Canvas visualization, live ledger with real click-throughs, and a MediaRecorder tap that exports a shareable audio clip of the master bus. A burst gate collapses event floods (a sniper sweep emitting dozens of identical guard events) into one accented note per actor per 1.5s. Solo mode narrows the stream to one participant client-side (`?agent=<id>` exact, `?actor=<label>` case-insensitive), reachable from the per-row "solo" control and from the "Listen →" link on every agent profile's Wallet Story card. No audio assets, no synthetic events. Tests: [tests/symphony-score.test.js](tests/symphony-score.test.js) || Platform activity feed | [pages/feed.html](pages/feed.html) `→ /feed` + [pages/community.html](pages/community.html) `→ /community`, backend [api/users/me/feed.js](api/users/me/feed.js) | Live | Reverse-chronological creation events (avatars, agents, coins, models, worlds) plus follow activity, merged at read time from `forge_creations`, `dioramas`, and `user_follows` (no separate event log to drift out of sync). `?scope=following` (default; session-gated, anonymous gets 401) vs `?scope=all` (public; powers `/community` and first-visit feeds); `limit` 1..50, `before` ISO cursor. Items deep-link to creator portfolios. Distinct from the Money Pulse ticker `GET /api/feed` ([docs/money-feed.md](docs/money-feed.md)). Docs: [docs/social-layer.md](docs/social-layer.md) || Contributor thank-you roll | [pages/contributors.html](pages/contributors.html) `→ /contributors`, linked from [pages/community.html](pages/community.html) and [README.md](README.md) | In-repo | A public acknowledgment page for people improving three.ws through code, reviews, issue reports, and design ideas. The code-contributor roll reads GitHub's public contributor record live, excludes bots and anonymous-email records, and has explicit loading, failure, and retry states; a curated section credits non-code contributions that led to concrete improvements, beginning with the Forge/R2 outage report in issue #164. |diff --git a/data/changelog.json b/data/changelog.jsonindex 7ae7eb24a..7ad10f476 100644--- a/data/changelog.json+++ b/data/changelog.json@@ -3,10 +3,17 @@"entries": [{"date": "2026-09-30",- "title": "Spotlight entries now show verified results, and success stories are opt-in",- "summary": "Every Agent Spotlight entry now has a Verified results block: coins the agent launched, creator fees it earned and income from other agents paying for its skills, computed by three.ws and linked to where you can check them. The builder cannot edit these numbers. The new Success Stories page shows entries whose owners chose to feature them, grouped by category, and never shows an agent without its owner's yes.",- "tags": ["feature"],- "link": "/stories"+ "title": "Experiment write-up: can autonomous agents trade new coins at a profit?",+ "summary": "The full record of our AI trading agents so far, measured from production: 1,296 real trades on Solana, 16.4% of them winners, and a net loss of 0.5402 SOL on 9.1444 SOL put to work. Only take-profit and take-initials exits made money, and the August retune made results worse, not better. The write-up says what that cost, what we got wrong, what we changed, and prints every query behind the numbers.",+ "tags": ["docs"],+ "link": "/docs/experiments-autonomous-trading"+ },+ {+ "date": "2026-09-30",+ "title": "Experiment write-up: how many new agents ever launch a coin?",+ "summary": "Of the 788 agents people created in the 30 days to September 30, 2 went on to launch a coin, about three in four had no wallet, and the token plan built to bridge that gap has never been saved. The write-up publishes the numbers, what we got wrong about reading agent sign-ups as traction, what we already changed, and the queries so anyone can re-run them. It is the first entry on the new experiments page.",+ "tags": ["docs"],+ "link": "/docs/experiments-agent-launch-funnel"},{"date": "2026-09-30",@@ -15,6 +22,13 @@"tags": ["feature"],"link": "/duels"},+ {+ "date": "2026-09-30",+ "title": "Spotlight entries now show verified results, and success stories are opt-in",+ "summary": "Every Agent Spotlight entry now has a Verified results block: coins the agent launched, creator fees it earned and income from other agents paying for its skills, computed by three.ws and linked to where you can check them. The builder cannot edit these numbers. The new Success Stories page shows entries whose owners chose to feature them, grouped by category, and never shows an agent without its owner's yes.",+ "tags": ["feature"],+ "link": "/stories"+ },{"date": "2026-09-30","title": "Sell your whole agent as a paid API",diff --git a/data/experiments.json b/data/experiments.jsonnew file mode 100644index 000000000..722aec1e3--- /dev/null+++ b/data/experiments.json@@ -0,0 +1,34 @@+{+ "$comment": "Public experiment log behind /experiments. One entry per experiment; the write-up lives at docs/<doc>.md and follows the template in docs/experiments.md. scripts/build-page-index.mjs validates every entry (npm run build:pages fails on a missing field or a missing write-up) and publishes public/experiments.json for the page. Newest first is not required; the page sorts by `published`.",+ "experiments": [+ {+ "slug": "agent-launch-funnel",+ "title": "How many new agents ever launch a coin?",+ "question": "Of the agents people create on three.ws, how many go on to launch a coin, and where does everyone else stop?",+ "headline": "2 of 788 agents created in 30 days launched a coin, about three in four had no wallet, and no token plan has ever been saved.",+ "published": "2026-09-30",+ "spend_usd": 0,+ "status": "concluded",+ "doc": "experiments-agent-launch-funnel",+ "related": [+ { "label": "Live funnel numbers", "href": "/analytics" },+ { "label": "Launch a coin", "href": "/launch" }+ ]+ },+ {+ "slug": "autonomous-trading",+ "title": "Can autonomous agents trade new coins at a profit?",+ "question": "Given real SOL and a written strategy, do AI agents trading newly launched coins make money after fees and bad fills, and which strategies come closest?",+ "headline": "1,296 real trades, 16.4% winners, a net loss of 0.5402 SOL on 9.1444 SOL deployed. Only take-profit and take-initials exits made money.",+ "published": "2026-09-30",+ "spend_usd": 63.87,+ "status": "running",+ "doc": "experiments-autonomous-trading",+ "live_url": "/sniper/experiments",+ "related": [+ { "label": "The experiment spec", "href": "/docs/trading-experiment" },+ { "label": "The first 90 trades, in depth", "href": "/blog/autonomous-trading-experiment" }+ ]+ }+ ]+}diff --git a/data/pages.json b/data/pages.jsonindex 86f8025df..0644e981f 100644--- a/data/pages.json+++ b/data/pages.json@@ -39,7 +39,7 @@},{"title": "Learn",- "paths": ["/docs", "/tutorials", "/examples", "/awesome", "/changelog", "/blog", "/status"]+ "paths": ["/docs", "/tutorials", "/examples", "/awesome", "/changelog", "/blog", "/status", "/experiments"]}]},@@ -2437,6 +2437,14 @@"changefreq": "daily","added": "2026-06-29"},+ {+ "path": "/experiments",+ "title": "Experiments",+ "description": "What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.",+ "priority": 0.7,+ "changefreq": "weekly",+ "added": "2026-09-30"+ },{"path": "/analytics","title": "Platform Analytics",@@ -5668,6 +5676,30 @@"changefreq": "monthly","added": "2026-07-30"},+ {+ "path": "/docs/experiments",+ "title": "Docs · How we run and publish experiments",+ "description": "The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.",+ "priority": 0.5,+ "changefreq": "monthly",+ "added": "2026-09-30"+ },+ {+ "path": "/docs/experiments-agent-launch-funnel",+ "title": "Experiment · How many new agents ever launch a coin?",+ "description": "Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.",+ "priority": 0.6,+ "changefreq": "monthly",+ "added": "2026-09-30"+ },+ {+ "path": "/docs/experiments-autonomous-trading",+ "title": "Experiment · Can autonomous agents trade new coins at a profit?",+ "description": "1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.",+ "priority": 0.6,+ "changefreq": "monthly",+ "added": "2026-09-30"+ },{"path": "/tutorials/getting-started","title": "Tutorial · Getting started (hosted)",diff --git a/docs/experiments-agent-launch-funnel.md b/docs/experiments-agent-launch-funnel.mdnew file mode 100644index 000000000..4d242eda9--- /dev/null+++ b/docs/experiments-agent-launch-funnel.md@@ -0,0 +1,148 @@+# Experiment: how many new agents ever launch a coin?++Published 2026-09-30. Status: concluded. Spend: $0. Index of every experiment: [/experiments](https://three.ws/experiments).++## The question++three.ws lets anyone create an AI agent and then launch a coin for it. A launched coin is where an agent starts to earn creator fees and where most of the platform's on-chain activity begins. So the question that decides whether the product works is not "how many agents are created" but: **of the agents people create, how many go on to launch a coin, and where does everyone else stop?**++A research pass on 2026-09-28 ([launchpad landscape](research/launchpad-landscape-2026-09.md)) reported 788 agents and 3 launches in 30 days. This write-up re-measures that funnel over a fixed, re-runnable window and adds the stages around it.++## Method++- **Window:** the 30 UTC days from 2026-08-31 00:00 to 2026-09-30 00:00 (end exclusive). The prior 30 days (2026-08-01 to 2026-08-31) are measured the same way for comparison. The window is pinned to fixed timestamps, not to `now()`, so re-running the queries later returns the same counts unless rows were hard-deleted since.+- **Tables:** `agent_identities` (every agent, including ones later soft-deleted), `pump_agent_mints` and `fixed_supply_launches` (every coin launched through three.ws), and `agent_token_plans` (the saved launch plan an agent can carry before it launches).+- **Network:** Solana mainnet unless a row says otherwise. Devnet rehearsals are counted on their own line.+- **Two figures are live state, not history.** An agent's wallet is provisioned when it first needs one, and an agent can be deleted later, so "holding a wallet" and "still live" describe those agents as they are when the query runs. The wallet count read 207 on the first run on 2026-09-30 and 208 on the re-run the same morning; it can only rise from here. Every other figure is fixed by the window.+- **Sample:** every row in the window, no sampling.+- **Tooling:** SELECT-only queries run against the production database on 2026-09-30. Every query is in the appendix.++## Spend++$0. This experiment is a measurement of what creators already did; nothing was bought, launched or paid for to run it.++## Result++| Stage, 2026-08-31 to 2026-09-30 | Count |+|---|---|+| Agents created | 788 |+| Distinct people who created them | 454 |+| Of those agents, still live (not deleted) | 744 |+| Of those agents, holding a Solana wallet address (live state, see Method) | 208 |+| Of those agents, launched a coin on mainnet (by 2026-09-30) | 2 |+| Of those agents, launched a devnet rehearsal coin | 0 |+| Coins launched on mainnet in the window (by any agent) | 3, from 3 agents and 3 people |+| Token plans ever saved (all time, any network) | 0 |++- **2 of 788 new agents launched a coin: 0.25%.** One of the three coins launched in the window came from an agent created before it.+- **208 of 788 new agents (26.4%) have a wallet address at all.** A coin launch signs with the agent's own custodial wallet, so roughly three in four new agents are not yet in a state where a launch could happen.+- **The token plan has never been saved.** The `agent_token_plans` table is empty. The object built to carry an agent from "created" to "launched" has not been used once.+- **Creation is growing while launches are not.** The prior 30 days had 509 agents from 315 people. Agents created rose 55% (509 to 788), while mainnet launches by calendar month went from 7 in August to 3 in September.++All-time context: 90 mainnet coins from 58 people and 65 agents since the first launch on 2026-05-13.++| Month | Mainnet coins | People who launched |+|---|---|---|+| 2026-05 | 26 | 5 |+| 2026-06 | 7 | 6 |+| 2026-07 | 47 | 38 |+| 2026-08 | 7 | 7 |+| 2026-09 (to 09-30) | 3 | 3 |++The re-measure differs slightly from the landscape doc (454 creators here, 452 there) because that pass used a rolling window ending when it was read on 2026-09-28, while this one uses the fixed window above.++## What we got wrong++- **We read agent creation as traction.** 788 agents a month looks like growth, and it is the number the home page shows. Measured against launches it is a 0.25% conversion, and launches are falling while creation rises. The number we celebrated was not the number that mattered.+- **We built the bridge and nobody crossed it.** The token plan shipped on 2026-08-11 (migration `20260811130000_agent_token_plans.sql`) to carry an agent from creation to launch. Zero plans have been saved. We did not measure whether creators ever reach the screen that saves one, so we cannot yet say which step loses them.+- **We did not know most new agents cannot launch.** Only 26.4% of new agents hold a wallet address, and a launch needs one. Nothing on any dashboard showed this until this measurement. While writing this we also found that our own [/analytics](https://three.ws/analytics) method note claimed every new agent gets a wallet; it was corrected the same day.+- **July was not the start of a trend.** 47 launches in July looked like momentum. August and September together produced 10.++## What we changed because of it++Changed already:++- **The funnel is now tracked daily in public.** [/analytics](https://three.ws/analytics) charts agents created, agents with a wallet and coins launched side by side over 30 days, 90 days or all time, with the method for each, so this ratio is watched instead of rediscovered.+- **A launch no longer has to start from a blank form.** `npx three-ws launch` hands a prefilled coin to `/launch` for the owner to sign ([CLI docs](cli.md)). It was built on 2026-09-29 and reaches creators once the `three-ws` package is published to npm, which is still pending.+- **Creators can see what a launched coin earns.** Per-agent creator earnings are now public on the agent profile and coin page, so the reason to launch is visible before launching.++Will change next, and re-measure with these same queries once it ships:++- **One default "launch my agent's coin" path** that drafts the token plan from the agent's identity and 3D render and asks for one confirmation (item 4 of the landscape doc's plan).+- **Instrument each step** between "agent created", "wallet provisioned", "plan saved" and "coin launched", so the next version of this write-up can say where people stop instead of only that they stop.++## Appendix: the queries++Every figure above comes from one of these SELECT statements, run against the production database on 2026-09-30. They use fixed timestamps, so anyone with read access can re-run them and get the same answer.++**Agents created in the window, the people who created them, and how many are live or hold a wallet** (788, 454, 744, 208):++```sql+select count(*) as agents_created, count(distinct user_id) as creating_users,+ count(*) filter (where deleted_at is null) as still_live,+ count(*) filter (where coalesce(meta->>'solana_address','') <> '') as with_solana_wallet+from agent_identities+where created_at >= '2026-08-31T00:00:00Z' and created_at < '2026-09-30T00:00:00Z';+```++**Of those agents, how many launched a coin** (2 mainnet, 0 devnet):++```sql+select count(distinct a.id) filter (where m.network = 'mainnet') as agents_launched_mainnet,+ count(distinct a.id) filter (where m.network = 'devnet') as agents_launched_devnet+from agent_identities a+join pump_agent_mints m on m.agent_id = a.id and m.created_at < '2026-09-30T00:00:00Z'+where a.created_at >= '2026-08-31T00:00:00Z' and a.created_at < '2026-09-30T00:00:00Z';+```++**Coins launched in the window, by network** (3 mainnet coins, 3 agents, 3 people):++```sql+select l.network, count(*) as coins, count(distinct l.agent_id) as agents, count(distinct l.user_id) as users+from (+ select network, agent_id, user_id, created_at from pump_agent_mints+ union all+ select network, agent_id, user_id, created_at from fixed_supply_launches+) l+where l.created_at >= '2026-08-31T00:00:00Z' and l.created_at < '2026-09-30T00:00:00Z'+group by l.network order by l.network;+```++**Token plans ever saved** (no rows, so 0):++```sql+select network, status, count(*) as plans,+ count(*) filter (where created_at >= '2026-08-31T00:00:00Z' and created_at < '2026-09-30T00:00:00Z') as plans_in_window+from agent_token_plans+where created_at < '2026-09-30T00:00:00Z'+group by network, status order by network, status;+```++**The prior 30 days, for comparison** (509 agents, 315 people):++```sql+select count(*) as agents_created, count(distinct user_id) as creating_users+from agent_identities+where created_at >= '2026-08-01T00:00:00Z' and created_at < '2026-08-31T00:00:00Z';+```++**Mainnet coins by month** (the monthly table):++```sql+select to_char(date_trunc('month', created_at at time zone 'UTC'), 'YYYY-MM') as month,+ count(*) as mainnet_coins, count(distinct user_id) as users+from pump_agent_mints+where network = 'mainnet' and created_at < '2026-09-30T00:00:00Z'+group by 1 order by 1;+```++**All-time mainnet coins** (90 coins, 58 people, 65 agents, first on 2026-05-13):++```sql+select count(*) as mainnet_coins, count(distinct user_id) as users, count(distinct agent_id) as agents,+ min(created_at) as first_launch+from pump_agent_mints+where network = 'mainnet' and created_at < '2026-09-30T00:00:00Z';+```++Derived figures: 2 / 788 = 0.25%; 208 / 788 = 26.4%; (788 - 509) / 509 = 55%.diff --git a/docs/experiments-autonomous-trading.md b/docs/experiments-autonomous-trading.mdnew file mode 100644index 000000000..fd0783781--- /dev/null+++ b/docs/experiments-autonomous-trading.md@@ -0,0 +1,210 @@+# Experiment: can autonomous agents trade new coins at a profit?++Published 2026-09-30. Status: running. Spend so far: $63.87 (0.5402 SOL of realized losses). Live scoreboard: [/sniper/experiments](https://three.ws/sniper/experiments). Index of every experiment: [/experiments](https://three.ws/experiments).++This is the standard write-up for the trading experiment. The full rule set, the exit math and how to run it live in [the experiment spec](trading-experiment.md); the long-form account of the first 90 trades, with charts, is the [blog post](https://three.ws/blog/autonomous-trading-experiment). This page adds the sections those two do not carry: the spend, the whole-history result measured on one date, and what it changed.++## The question++Give AI agents their own Solana wallets and real SOL, let each one trade newly launched coins under a written strategy (fixed rules, or a language model judging each coin), and record why every buy and sell happened. **After fees and bad fills, do they make money, and which strategies come closest?**++## Method++- **Window:** every real mainnet trade from the first fill on 2026-07-03 to 2026-09-30 00:00 UTC (end exclusive, pinned so the queries re-run to the same answer).+- **Tables:** `agent_sniper_strategies` (each armed strategy, its rules and whether it is judged by rules or a model), `agent_sniper_positions` (each trade: entry, exit, realized P&L), `sniper_llm_verdicts` (every model judgment, bought or not).+- **Real fills only:** a position counts when it has an on-chain buy signature. Paper trades from simulate mode carry the `SIMULATED` sentinel and are reported on their own line, never mixed in.+- **Sample:** 28 strategies on 28 agents in 4 experiment groups, 1,296 closed trades. No sampling.+- **Tools:** the sniper worker (`workers/agent-sniper/`), the arm API (`api/sniper/strategy.js`), and SELECT-only queries run against the production database on 2026-09-30. Every query is in the appendix.++## Spend++**0.5402 SOL of net realized losses, $63.87** valued at the live SOL price when this was measured ($118.23 at 2026-09-30 08:25 UTC, read through the platform's own price helper). That is the money the experiment has cost so far. It is valued at one price on one date, not at the price of each trade.++Two costs are listed but not added, because we could not attribute them honestly:++- **Tips:** positions record 0.001035 SOL of priority tips in `tip_lamports`. We have not verified whether realized P&L already nets them, so they are shown, not summed.+- **Model calls:** the judges produced 357,988 mainnet verdicts since 2026-07-21. The verdict table carries no cost column, so the language-model cost of the experiment is not measured.++## Result++| Whole history, mainnet, real fills | Value |+|---|---|+| Closed trades | 1,296 |+| Winners (realized P&L above zero) | 212 (16.4%) |+| SOL deployed across closed trades | 9.1444 SOL |+| Net realized P&L | -0.5402 SOL (-5.91% of deployed) |+| Open positions on 2026-09-30 | 0 |+| Paper (simulate-mode) trades closed | 15, 1 winner |++**By how the strategy decides:**++| Decision mode | Closed | Winners | Net realized | Of deployed |+|---|---|---|---|---|+| Fixed rules | 1,220 | 196 (16.1%) | -0.5196 SOL | -6.17% |+| Language-model judge | 76 | 16 (21.1%) | -0.0206 SOL | -2.86% |++**How trades ended:**++| Exit reason | Trades | Net realized |+|---|---|---|+| Liquidity decay (nobody trading the coin) | 468 | -0.1257 SOL |+| Stop-loss | 318 | -0.6752 SOL |+| Trailing stop | 297 | -0.2304 SOL |+| Timeout | 140 | -0.0602 SOL |+| Take-profit | 67 | +0.5512 SOL |+| Error | 6 | 0 SOL |++**Before and after the 2026-08-09 retune:**++| Period (by trade open) | Closed | Winners | Net realized | Of deployed |+|---|---|---|---|---|+| Before 2026-08-09 | 241 | 80 (33.2%) | -0.0594 SOL | -1.13% |+| From 2026-08-09 | 1,055 | 132 (12.5%) | -0.4808 SOL | -12.37% |++**The take-initials ladder** (sell enough at 2x to recover the stake, keep the rest): the 63 trades where it fired made +0.5298 SOL. The other 1,233 lost 1.0700 SOL.++**By month:**++| Month | Closed | Winners | Net realized |+|---|---|---|---|+| 2026-07 | 241 | 80 | -0.0594 SOL |+| 2026-08 | 437 | 39 | -0.2842 SOL |+| 2026-09 | 618 | 93 | -0.1966 SOL |++What the numbers say: the fleet as a whole loses money. The only exits that make money are the take-profit and take-initials paths, and they are rare. The language-model arms lost less per SOL deployed than the rule arms, on a sample one sixteenth the size, which is not yet enough to call. The judges said buy on 16,860 of 357,988 verdicts (4.7%).++## What we got wrong++- **The retune made the numbers worse, not better.** The 2026-08-09 retune (section "The 2026-08-09 retune" of [the spec](trading-experiment.md)) rewrote the strategy table around what the first trades said. Measured by trade open date, the win rate after it fell from 33.2% to 12.5%, and the loss per SOL deployed grew from 1.13% to 12.37%. The fleet also traded far more after it (241 closed trades before, 1,055 after). Either way, it did not do what it was for.+- **Stops do not cap the downside.** Stop-loss exits alone lost 0.6752 SOL, more than the whole fleet's net loss. The spec already records why: on a thin bonding curve a coin can gap straight through a stop before it fires.+- **The trailing stop, meant to lock in gains, lost money in aggregate** (297 trades, -0.2304 SOL). We have not yet compared each trade's peak value with its exit value, which is the measurement that would say why.+- **We measured win rate first.** Early reporting led with win rate. It is the wrong headline: the rule arms won 16.1%, and the only exit groups that made money are the 67 take-profits (+0.5512 SOL) and the 63 ladder recoveries (+0.5298 SOL), which can overlap. The blog post makes the same case in its section "Win rate is a vanity metric".+- **We cannot state the full cost.** The model calls behind 357,988 verdicts were never priced per verdict, so "spend" above is trading losses only.++## What we changed because of it++Changed already, each documented in [the spec](trading-experiment.md):++- **The take-initials ladder is enforced on every enabled arm**, because it is the one exit pattern that made money, and the fleet never sells the whole of a winning position.+- **A liquidity-decay exit** closes an underwater position whose price has frozen, instead of letting it sit until the timeout. It now ends more trades than any other exit (468).+- **A confidence ceiling and strict-model mode for the model arms**, so an over-confident verdict or a fallback model's answer is recorded but not funded.+- **A row-count watchdog** (`/api/cron/sniper-loops-health`) that pages when any learning loop stops writing rows, and a `stall` field on the live scoreboard naming why an arm is not trading.++Will change next, and re-measure with the same queries:++- **Size entries as if every coin can go to zero**, since stops do not hold. The next measurement should show stop-loss losses shrinking as a share of deployed SOL.+- **Price the model calls**, so the spend line can include the judges and the model arms can be compared on cost as well as P&L.+- **Judge the retune by the numbers above**, not by the reasoning behind it: if the post-retune loss rate does not fall in the next window, roll the strategy table back toward the pre-retune shape.++## Appendix: the queries++Every figure above comes from one of these SELECT statements, run against the production database on 2026-09-30, with fixed end timestamps so they re-run to the same answer. The SOL price is the one exception: it was read once, live, through `solPriceUsd()` in `api/_lib/sol-price.js` at 2026-09-30 08:25 UTC and returned 118.23.++**Strategies** (28 strategies, 28 agents, 4 experiment groups; 11 enabled on 2026-09-30; first armed 2026-06-18):++```sql+select count(*) as strategies, count(*) filter (where enabled) as enabled_now,+ count(distinct experiment_group) as groups, count(distinct agent_id) as agents,+ min(created_at) as first_armed+from agent_sniper_strategies+where network = 'mainnet' and created_at < '2026-09-30T00:00:00Z';+```++**Whole-history result** (1,296 closed; 212 winners; 0 open; -540,179,418 lamports realized on 9,144,352,384 deployed; 1,035,000 lamports of tips; first fill 2026-07-03):++```sql+select count(*) filter (where status = 'closed') as closed,+ count(*) filter (where status = 'closed' and realized_pnl_lamports > 0) as wins,+ count(*) filter (where status = 'open') as open_now,+ coalesce(sum(realized_pnl_lamports) filter (where status = 'closed'), 0) as realized_pnl_lamports,+ coalesce(sum(entry_quote_lamports) filter (where status = 'closed'), 0) as deployed_lamports,+ coalesce(sum(tip_lamports), 0) as tips_lamports,+ min(opened_at) as first_trade, max(closed_at) as last_close+from agent_sniper_positions+where network = 'mainnet' and buy_sig is not null and buy_sig <> 'SIMULATED'+ and opened_at < '2026-09-30T00:00:00Z';+```++**By decision mode** (the rules and model table):++```sql+select coalesce(s.decision_mode, 'rules') as mode,+ count(*) filter (where p.status = 'closed') as closed,+ count(*) filter (where p.status = 'closed' and p.realized_pnl_lamports > 0) as wins,+ coalesce(sum(p.realized_pnl_lamports) filter (where p.status = 'closed'), 0) as realized_pnl_lamports,+ coalesce(sum(p.entry_quote_lamports) filter (where p.status = 'closed'), 0) as deployed_lamports+from agent_sniper_positions p+join agent_sniper_strategies s on s.id = p.strategy_id+where p.network = 'mainnet' and p.buy_sig is not null and p.buy_sig <> 'SIMULATED'+ and p.opened_at < '2026-09-30T00:00:00Z'+group by 1 order by 1;+```++**By exit reason** (the exit table):++```sql+select coalesce(exit_reason, 'none') as exit_reason, count(*) as closed,+ coalesce(sum(realized_pnl_lamports), 0) as realized_pnl_lamports+from agent_sniper_positions+where network = 'mainnet' and status = 'closed' and buy_sig is not null and buy_sig <> 'SIMULATED'+ and opened_at < '2026-09-30T00:00:00Z'+group by 1 order by closed desc;+```++**Paper trades** (15 closed, 1 winner):++```sql+select count(*) filter (where status = 'closed') as paper_closed,+ count(*) filter (where status = 'closed' and realized_pnl_lamports > 0) as paper_wins+from agent_sniper_positions+where network = 'mainnet' and buy_sig = 'SIMULATED' and opened_at < '2026-09-30T00:00:00Z';+```++**By month** (the monthly table):++```sql+select to_char(date_trunc('month', opened_at at time zone 'UTC'), 'YYYY-MM') as month,+ count(*) filter (where status = 'closed') as closed,+ count(*) filter (where status = 'closed' and realized_pnl_lamports > 0) as wins,+ coalesce(sum(realized_pnl_lamports) filter (where status = 'closed'), 0) as realized_pnl_lamports+from agent_sniper_positions+where network = 'mainnet' and buy_sig is not null and buy_sig <> 'SIMULATED'+ and opened_at < '2026-09-30T00:00:00Z'+group by 1 order by 1;+```++**Model verdicts** (357,988 verdicts, 16,860 buys, first on 2026-07-21):++```sql+select count(*) as verdicts, count(*) filter (where buy) as said_buy, min(created_at) as first_verdict+from sniper_llm_verdicts+where network = 'mainnet' and created_at < '2026-09-30T00:00:00Z';+```++**Before and after the retune** (the retune table):++```sql+select case when opened_at < '2026-08-09T00:00:00Z' then 'before 2026-08-09' else 'from 2026-08-09' end as period,+ count(*) filter (where status = 'closed') as closed,+ count(*) filter (where status = 'closed' and realized_pnl_lamports > 0) as wins,+ coalesce(sum(realized_pnl_lamports) filter (where status = 'closed'), 0) as realized_pnl_lamports,+ coalesce(sum(entry_quote_lamports) filter (where status = 'closed'), 0) as deployed_lamports+from agent_sniper_positions+where network = 'mainnet' and buy_sig is not null and buy_sig <> 'SIMULATED'+ and opened_at < '2026-09-30T00:00:00Z'+group by 1 order by 1;+```++**The take-initials ladder** (63 trades +529,834,054 lamports; 1,233 trades -1,070,013,472 lamports):++```sql+select coalesce(initials_recovered, false) as initials_recovered,+ count(*) filter (where status = 'closed') as closed,+ coalesce(sum(realized_pnl_lamports) filter (where status = 'closed'), 0) as realized_pnl_lamports+from agent_sniper_positions+where network = 'mainnet' and buy_sig is not null and buy_sig <> 'SIMULATED'+ and opened_at < '2026-09-30T00:00:00Z'+group by 1 order by 1;+```++Derived figures: 1 SOL = 1,000,000,000 lamports. 212 / 1,296 = 16.4%; -0.5402 / 9.1444 = -5.91%; 196 / 1,220 = 16.1%; -0.5196 / 8.4249 = -6.17%; 16 / 76 = 21.1%; -0.0206 / 0.7195 = -2.86%; 80 / 241 = 33.2%; -0.0594 / 5.2567 = -1.13%; 132 / 1,055 = 12.5%; -0.4808 / 3.8876 = -12.37%; 16,860 / 357,988 = 4.7%; 0.540179418 x 118.23 = $63.87.diff --git a/docs/experiments.md b/docs/experiments.mdnew file mode 100644index 000000000..fd7f51155--- /dev/null+++ b/docs/experiments.md@@ -0,0 +1,49 @@+# Experiments: how we publish what we tried++[/experiments](https://three.ws/experiments) is the public log of what three.ws has tried, what it cost, what the numbers said, and what we got wrong. Every entry is measured from production data, and every number in a write-up comes with the query that produced it, so anyone can check it.++## What is on the page++Each card is one experiment: the question it asked, its status, what it cost in dollars, and the date the write-up was published. The status filter shows running, concluded or killed experiments. A card opens its full write-up, and when an experiment has a live view (for example the trading scoreboard at [/sniper/experiments](https://three.ws/sniper/experiments)) the card links to that as well.++Current write-ups:++- [How many new agents ever launch a coin?](experiments-agent-launch-funnel.md)+- [Can autonomous agents trade new coins at a profit?](experiments-autonomous-trading.md)++## Adding an experiment++1. **Measure first.** Run SELECT-only queries (or a script that only reads) against production data. Pin every time window to fixed timestamps rather than `now()`, so the queries re-run to the same answer later. Never estimate or round a number to make it look better; if something cannot be measured, say "not measured" and why.+2. **Write the write-up** at `docs/experiments-<slug>.md`, with these six sections as `## ` headings, in this order:++ | Section | What goes in it |+ |---|---|+ | The question | One question the experiment answers, and why it matters. |+ | Method | The data window, the tables or tools, the sample size, and what was excluded. |+ | Spend | What the experiment cost, in dollars, and how it was valued. $0 is a valid answer. |+ | Result | The numbers, in tables where possible, and what they say. |+ | What we got wrong | The unflattering part. Publish it. |+ | What we changed because of it | What already changed, and what will change next. |++ Close with an `## Appendix: the queries` section holding every query that produced a number above it, plus the arithmetic for any derived figure.+3. **Add an entry** to [data/experiments.json](https://github.com/nirholas/three.ws/blob/main/data/experiments.json):++ ```json+ {+ "slug": "agent-launch-funnel",+ "title": "How many new agents ever launch a coin?",+ "question": "Of the agents people create on three.ws, how many go on to launch a coin, and where does everyone else stop?",+ "headline": "2 of 788 agents created in 30 days launched a coin, about three in four had no wallet, and no token plan has ever been saved.",+ "published": "2026-09-30",+ "spend_usd": 0,+ "status": "concluded",+ "doc": "experiments-agent-launch-funnel",+ "related": [{ "label": "Live funnel numbers", "href": "/analytics" }]+ }+ ```++ Required: `slug` (lowercase kebab-case), `title`, `question`, `published` (YYYY-MM-DD), `spend_usd` (a number, 0 or more), `status` (`running`, `concluded` or `killed`) and `doc` (always `experiments-<slug>`). Optional: `headline` (the result in one sentence), `live_url` (a site path to a live view) and `related` (extra site links).+4. **Register the page.** Add `/docs/experiments-<slug>` to `data/pages.json`, a link under "Experiments" in `docs/nav.json`, and a `data/changelog.json` entry announcing the write-up.+5. **Build.** `npm run build:pages` validates every entry and writes `public/experiments.json`, which the page reads. It fails on a missing field, an unknown status, a `doc` that does not exist, or a write-up that is missing one of the six sections or has them out of order.++The validator lives in [scripts/lib/experiments-index.mjs](https://github.com/nirholas/three.ws/blob/main/scripts/lib/experiments-index.mjs) and is covered by `tests/experiments-index.test.js`.diff --git a/docs/nav.json b/docs/nav.jsonindex a735a57b5..17c40fc28 100644--- a/docs/nav.json+++ b/docs/nav.json@@ -487,6 +487,23 @@}]},+ {+ "title": "Experiments",+ "links": [+ {+ "label": "How we run and publish experiments",+ "path": "experiments"+ },+ {+ "label": "How many new agents ever launch a coin?",+ "path": "experiments-agent-launch-funnel"+ },+ {+ "label": "Can autonomous agents trade at a profit?",+ "path": "experiments-autonomous-trading"+ }+ ]+ },{"title": "Getting started","links": [diff --git a/docs/start-here.md b/docs/start-here.mdindex 1d0bc38ff..1724e874b 100644--- a/docs/start-here.md+++ b/docs/start-here.md@@ -138,6 +138,7 @@ Want the model in your hands rather than on a screen? [Materialize](./materializ- **Want other agents to pay for your tools?** → [Monetize an MCP server](./tutorials/monetize-mcp-server.md): build an MCP server that charges per `tools/call` in USDC on Solana, with a free orientation tool in front of the paid ones and the verify-work-settle ordering that means a failed call never takes money. The runnable code is [examples/paid-mcp-server](https://github.com/nirholas/three.ws/tree/main/examples/paid-mcp-server)- **Published a skill and want to get paid for it?** → [Skill royalties](./skill-royalties.md): every paid call to a skill you wrote routes your share straight to your own wallet at settlement (Solana first), with the exact split, the ledger where you watch it land with an explorer link per payout, and a proof lane that verifies the whole accounting path without moving a cent- **Want the whole platform in one view?** → [Platform analytics](https://three.ws/analytics) puts every platform total on one page (agents, wallets, coins launched, 3D models, LLM tokens, x402 settlements, marketplace sales, agent-to-agent hire volume), each with a daily growth chart and a note saying exactly which rows were counted. The same data is open at `GET /api/platform/analytics` ([API reference](./api-reference.md#platform-analytics-api))+- **Want to know what we tried and what failed?** → [Experiments](https://three.ws/experiments) are our public write-ups: the question, the method, what it cost, the result measured from production data, what we got wrong and what we changed, with every query printed so you can re-run it. Start with [how many new agents ever launch a coin](./experiments-agent-launch-funnel.md) and [whether autonomous agents can trade at a profit](./experiments-autonomous-trading.md); [how we publish one](./experiments.md) covers the format- **Checking whether the economy is real?** → [Agent economy volume](./agent-economy-volume.md) counts only settled agent-to-agent USDC for paid skills, and [Viability](./viability.md) shows the unit economics (GMV, take rate, repeat buyers, realized P&L) from on-chain data with no projections- **Designing an agent's look and brand?** → [Agent Identity Studio](./agent-identities.md) turns a brand brief into a rigged avatar plus posed studio renders; [Living Stages](./stage.md) covers shareable stages; [the Avatar Engines Atlas](./avatar-engines.md) is the factual survey of the engines in this space and which ones we animate- **Need to prove what an agent was told to do?** → [Signed manifests](./agent-manifest.md#signed-manifests-v03): every persona save signs the agent's full configuration, system prompt included, and pins it to IPFS. Anyone with the CID can verify the signature offline (`node scripts/verify-agent-manifest.mjs --cid …`) and diff the pinned record against the agent running today, with no account and no trust in usdiff --git a/pages/experiments.html b/pages/experiments.htmlnew file mode 100644index 000000000..155bd1adb--- /dev/null+++ b/pages/experiments.html@@ -0,0 +1,143 @@+<!DOCTYPE html>+<html lang="en">+<head>+ <style>/* three.ws hidden-guard: make the `hidden` attribute authoritative; see scripts/lib/hidden-guard.mjs */[hidden]{display:none!important}</style>+ <script>/* three.ws theme boot: no-flash */(function(){try{var m=localStorage.getItem('twx_theme');var l=m==='auto'?(window.matchMedia&&window.matchMedia('(prefers-color-scheme: light)').matches):m==='light';document.documentElement.setAttribute('data-theme',l?'light':'dark');}catch(e){document.documentElement.setAttribute('data-theme','dark');}})();</script>+ <meta charset="utf-8" />+ <meta name="viewport" content="width=device-width, initial-scale=1, viewport-fit=cover" />+ <title>Experiments · three.ws</title>+ <meta name="description" content="What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ships the queries that produced its numbers." />+ <link rel="canonical" href="https://three.ws/experiments" />+ <link rel="icon" href="/favicon.ico" sizes="any" />+ <link rel="icon" type="image/svg+xml" href="/favicon.svg" />+ <link rel="apple-touch-icon" href="/apple-touch-icon.png" />+ <link rel="manifest" href="/manifest.webmanifest" />+ <link rel="stylesheet" href="/style.css" />+ <link rel="stylesheet" href="/nav.css" />+ <link rel="stylesheet" href="/mobile.css" />+ <meta property="og:type" content="website" />+ <meta property="og:site_name" content="three.ws" />+ <meta property="og:title" content="three.ws Experiments" />+ <meta property="og:description" content="What we tried, what it cost, what the numbers said and what we got wrong, with the queries behind every number." />+ <meta property="og:url" content="https://three.ws/experiments" />+ <meta property="og:image" content="https://three.ws/api/page-og?s=learn&t=Experiments&d=What+three.ws+tried%2C+what+it+cost%2C+what+the+numbers+said+and+what+we+got+wrong%2C+with+the+queries+behind+every+number.&p=%2Fexperiments" />+ <meta property="og:image:alt" content="three.ws Experiments" />+ <meta property="og:locale" content="en_US" />+ <meta name="twitter:card" content="summary_large_image" />+ <meta name="twitter:site" content="@trythreews" />+ <meta name="twitter:title" content="three.ws Experiments" />+ <meta name="twitter:description" content="What we tried, what it cost, what the numbers said and what we got wrong." />+ <meta name="twitter:image" content="https://three.ws/api/page-og?s=learn&t=Experiments&d=What+three.ws+tried%2C+what+it+cost%2C+what+the+numbers+said+and+what+we+got+wrong%2C+with+the+queries+behind+every+number.&p=%2Fexperiments" />+ <meta name="robots" content="index, follow, max-image-preview:large, max-snippet:-1" />+ <script type="application/ld+json">{"@context":"https://schema.org","@type":"CollectionPage","name":"Experiments","description":"What three.ws tried, what it cost, what the numbers said and what we got wrong.","url":"https://three.ws/experiments","isPartOf":{"@type":"WebSite","name":"three.ws","url":"https://three.ws"}}</script>+ <style>+ :root { --ex-maxw: 980px; --ex-run: #16a34a; --ex-kill: var(--danger, #f87171); }+ :root:not([data-theme="light"]) { --ex-run: #4ade80; }+ * { box-sizing: border-box; }+ html, body { margin: 0; background: var(--bg-0); color: var(--ink); }+ body { font-family: var(--font-body, 'Inter', system-ui, sans-serif); min-height: 100vh; }+ .site-header { position: sticky; top: 0; z-index: 50; background: color-mix(in srgb, var(--bg-0) 82%, transparent); backdrop-filter: blur(10px); }+ a { color: inherit; }++ .ex-wrap { max-width: var(--ex-maxw); margin: 0 auto; padding: 0 18px 96px; }+ .ex-hero { padding: 40px 0 6px; }+ .ex-eyebrow { font-size: 0.74rem; text-transform: uppercase; letter-spacing: 0.08em; color: var(--ink-dim); font-weight: 650; margin: 0 0 10px; }+ .ex-hero h1 { font-size: clamp(1.8rem, 4.6vw, 2.7rem); line-height: 1.1; margin: 0 0 12px; letter-spacing: -0.025em; color: var(--ink-bright, var(--ink)); }+ .ex-hero p { margin: 0 0 10px; max-width: 66ch; color: var(--ink-dim); line-height: 1.6; font-size: 1.02rem; }+ .ex-hero a { text-decoration: underline; text-underline-offset: 3px; }+ .ex-hero a:hover { color: var(--ink-bright, var(--ink)); }+ .ex-hero a:focus-visible, .ex-card a:focus-visible, .ex-empty a:focus-visible { outline: 2px solid var(--accent); outline-offset: 2px; border-radius: 4px; }++ .ex-filter { display: inline-flex; flex-wrap: wrap; gap: 2px; margin: 22px 0 18px; background: var(--surface-1); border: 1px solid var(--stroke); border-radius: 10px; padding: 3px; }+ .ex-filter button {+ appearance: none; border: 0; background: none; color: var(--ink-dim); font: inherit; font-size: 0.84rem; font-weight: 600;+ padding: 6px 12px; min-height: 32px; border-radius: 7px; cursor: pointer; display: inline-flex; align-items: center; gap: 6px;+ transition: background var(--duration-fast, 140ms) var(--ease-standard, ease), color var(--duration-fast, 140ms) var(--ease-standard, ease), transform 80ms;+ }+ .ex-filter button:hover { color: var(--ink-bright, var(--ink)); background: var(--surface-2); }+ .ex-filter button:active { transform: translateY(1px); }+ .ex-filter button:focus-visible { outline: 2px solid var(--accent); outline-offset: 1px; }+ .ex-filter button[aria-pressed="true"] { background: var(--surface-3); color: var(--ink-bright, var(--ink)); }+ .ex-count { font-size: 0.72rem; font-variant-numeric: tabular-nums; opacity: 0.75; }++ .ex-list { display: grid; gap: 12px; }+ .ex-card {+ padding: 20px 20px 18px; border-radius: var(--radius-md, 10px); background: var(--surface-1); border: 1px solid var(--stroke);+ transition: border-color var(--duration-fast, 140ms) var(--ease-standard, ease), background var(--duration-fast, 140ms) var(--ease-standard, ease);+ animation: ex-in 240ms var(--ease-standard, ease) both;+ }+ .ex-card:hover { border-color: var(--stroke-strong); background: var(--surface-2); }+ .ex-meta { display: flex; flex-wrap: wrap; align-items: center; gap: 8px 14px; font-size: 0.8rem; color: var(--ink-dim); font-variant-numeric: tabular-nums; }+ .ex-status { display: inline-flex; align-items: center; gap: 6px; font-weight: 650; padding: 3px 9px 3px 8px; border-radius: 999px; border: 1px solid var(--stroke-strong); color: var(--ink); }+ .ex-status::before { content: ''; width: 7px; height: 7px; border-radius: 50%; background: currentColor; }+ .ex-status[data-status="running"] { color: var(--ex-run); border-color: color-mix(in srgb, var(--ex-run) 45%, transparent); }+ .ex-status[data-status="killed"] { color: var(--ex-kill); border-color: color-mix(in srgb, var(--ex-kill) 45%, transparent); }+ .ex-status[data-status="concluded"] { color: var(--ink-dim); }+ .ex-card h2 { margin: 12px 0 6px; font-size: 1.2rem; letter-spacing: -0.01em; line-height: 1.3; }+ .ex-card h2 a { text-decoration: none; color: var(--ink-bright, var(--ink)); }+ .ex-card h2 a:hover { text-decoration: underline; text-underline-offset: 3px; }+ .ex-q { margin: 0; color: var(--ink); line-height: 1.55; font-size: 0.95rem; }+ .ex-q strong { color: var(--ink-dim); font-weight: 600; }+ .ex-headline { margin: 10px 0 0; padding-left: 12px; border-left: 2px solid var(--stroke-strong); color: var(--ink-dim); line-height: 1.55; font-size: 0.92rem; }+ .ex-links { display: flex; flex-wrap: wrap; gap: 8px 16px; margin-top: 14px; font-size: 0.88rem; }+ .ex-links a { text-decoration: underline; text-underline-offset: 3px; color: var(--ink-dim); }+ .ex-links a:hover { color: var(--ink-bright, var(--ink)); }+ .ex-links a.is-primary { color: var(--ink-bright, var(--ink)); font-weight: 650; }++ .ex-skel { height: 170px; border-radius: var(--radius-md, 10px); }+ .skeleton { background: var(--surface-2); background-image: linear-gradient(90deg, var(--surface-1) 0%, var(--surface-3) 50%, var(--surface-1) 100%); background-size: 200% 100%; animation: ex-shimmer 1.4s ease-in-out infinite; }+ @keyframes ex-shimmer { 0% { background-position: 200% 0; } 100% { background-position: -200% 0; } }+ @keyframes ex-in { from { opacity: 0; transform: translateY(4px); } to { opacity: 1; transform: none; } }++ .ex-empty { padding: 28px 20px; border-radius: var(--radius-md, 10px); border: 1px dashed var(--stroke-strong); color: var(--ink-dim); line-height: 1.6; }+ .ex-empty strong { color: var(--ink-bright, var(--ink)); display: block; margin-bottom: 4px; }+ .ex-empty a, .ex-empty button { color: var(--ink-bright, var(--ink)); text-decoration: underline; text-underline-offset: 3px; }+ .ex-empty button { appearance: none; background: none; border: 0; font: inherit; padding: 0; cursor: pointer; }+ .ex-empty button:focus-visible { outline: 2px solid var(--accent); outline-offset: 2px; border-radius: 4px; }+ .ex-error { display: flex; align-items: center; justify-content: space-between; gap: 14px; flex-wrap: wrap; padding: 14px 16px; border-radius: var(--radius-md, 10px); border: 1px solid color-mix(in srgb, var(--danger) 45%, transparent); background: color-mix(in srgb, var(--danger) 9%, transparent); line-height: 1.5; font-size: 0.9rem; }+ .ex-error button { appearance: none; border: 1px solid var(--stroke-strong); background: var(--surface-2); color: var(--ink-bright, var(--ink)); font: inherit; font-size: 0.85rem; font-weight: 600; padding: 7px 14px; border-radius: 8px; cursor: pointer; transition: background var(--duration-fast, 140ms), transform 80ms; }+ .ex-error button:hover { background: var(--surface-3); }+ .ex-error button:active { transform: translateY(1px); }+ .ex-error button:focus-visible { outline: 2px solid var(--accent); outline-offset: 2px; }++ .ex-foot { margin-top: 28px; font-size: 0.86rem; color: var(--ink-dim); line-height: 1.6; }+ .ex-foot a { text-decoration: underline; text-underline-offset: 3px; }+ .ex-foot a:hover { color: var(--ink-bright, var(--ink)); }+ .ex-foot a:focus-visible { outline: 2px solid var(--accent); outline-offset: 2px; border-radius: 4px; }++ @media (prefers-reduced-motion: reduce) { .skeleton, .ex-card { animation: none; } }+ </style>+</head>+<body>+ <header class="site-header"><div id="nav-container"></div></header>+ <script src="/nav.js"></script>++ <main class="ex-wrap" id="main">+ <section class="ex-hero">+ <p class="ex-eyebrow">Experiments</p>+ <h1>What we tried, and what the numbers said</h1>+ <p>Each write-up asks one question, says what it cost, shows the result measured from production data, and names what we got wrong and what we changed because of it. The queries behind every number are printed at the end, so you can re-run them.</p>+ <p><a href="/docs/experiments">How we run and publish an experiment</a></p>+ </section>++ <div class="ex-filter" id="ex-filter" role="group" aria-label="Filter by status">+ <button type="button" data-status="all" aria-pressed="true">All <span class="ex-count" data-count="all"></span></button>+ <button type="button" data-status="running" aria-pressed="false">Running <span class="ex-count" data-count="running"></span></button>+ <button type="button" data-status="concluded" aria-pressed="false">Concluded <span class="ex-count" data-count="concluded"></span></button>+ <button type="button" data-status="killed" aria-pressed="false">Killed <span class="ex-count" data-count="killed"></span></button>+ </div>++ <div class="ex-list" id="ex-list" aria-live="polite" aria-busy="true">+ <div class="ex-skel skeleton" aria-hidden="true"></div>+ <div class="ex-skel skeleton" aria-hidden="true"></div>+ </div>++ <p class="ex-foot">Numbers that move every day live on <a href="/analytics">platform analytics</a>. The trading experiment's live scoreboard is <a href="/sniper/experiments">/sniper/experiments</a>.</p>+ </main>++ <div id="footer-container"></div>+ <script src="/footer.js"></script>+ <script type="module" src="/src/experiments-index.js"></script>+ <script type="module" src="/i18n.js"></script>+</body>+</html>diff --git a/public/changelog-recent.json b/public/changelog-recent.jsonindex 20cd794f9..08d35bee4 100644--- a/public/changelog-recent.json+++ b/public/changelog-recent.json@@ -1,14 +1,64 @@{- "generated_at": "2026-09-30T08:30:10.238Z",+ "generated_at": "2026-09-30T08:35:45.908Z","generated_by": "scripts/build-page-index.mjs from data/pages.json + data/changelog.json","site": {"name": "three.ws","url": "https://three.ws"},- "total": 3794,+ "total": 3802,"complete": false,"full_feed": "/changelog.json","entries": [+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Docs · How we run and publish experiments",+ "summary": "The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.",+ "link": "/docs/experiments",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Docs · Trader duels",+ "summary": "How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.",+ "link": "/docs/trader-duels",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Experiment · Can autonomous agents trade new coins at a profit?",+ "summary": "1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.",+ "link": "/docs/experiments-autonomous-trading",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Experiment · How many new agents ever launch a coin?",+ "summary": "Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.",+ "link": "/docs/experiments-agent-launch-funnel",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Experiments",+ "summary": "What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.",+ "link": "/experiments",+ "tags": [+ "launch"+ ]+ },{"date": "2026-09-30","type": "launch",@@ -70,6 +120,26 @@"feature"]},+ {+ "date": "2026-09-30",+ "type": "update",+ "title": "Experiment write-up: can autonomous agents trade new coins at a profit?",+ "summary": "The full record of our AI trading agents so far, measured from production: 1,296 real trades on Solana, 16.4% of them winners, and a net loss of 0.5402 SOL on 9.1444 SOL put to work. Only take-profit and take-initials exits made money, and the August retune made results worse, not better. The write-up says what that cost, what we got wrong, what we changed, and prints every query behind the numbers.",+ "link": "/docs/experiments-autonomous-trading",+ "tags": [+ "docs"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "update",+ "title": "Experiment write-up: how many new agents ever launch a coin?",+ "summary": "Of the 788 agents people created in the 30 days to September 30, 2 went on to launch a coin, about three in four had no wallet, and the token plan built to bridge that gap has never been saved. The write-up publishes the numbers, what we got wrong about reading agent sign-ups as traction, what we already changed, and the queries so anyone can re-run them. It is the first entry on the new experiments page.",+ "link": "/docs/experiments-agent-launch-funnel",+ "tags": [+ "docs"+ ]+ },{"date": "2026-09-30","type": "update",@@ -152,6 +222,16 @@"improvement"]},+ {+ "date": "2026-09-30",+ "type": "update",+ "title": "Trader duels: call which trader wins, free to play",+ "summary": "A new Duels page pairs traders who sit next to each other on the board and asks one question: who books more realized profit over tomorrow, or over next week? Make your call before the window starts with free points from your daily allowance, then watch the race live on the duel page with both traders' records, trade rooms and profiles one tap away. Results come straight from the on-chain trade ledger, the same numbers their profiles show, and ties or empty windows refund every call. A correct call doubles your stake and adds XP to your quest level, and there is a monthly season board and four badges to earn. Points have no cash value and cannot be bought, sold, transferred or redeemed.",+ "link": "/duels",+ "tags": [+ "feature"+ ]+ },{"date": "2026-09-29","type": "launch",@@ -1497,93 +1577,6 @@"tags": ["fix"]- },- {- "date": "2026-09-09",- "type": "update",- "title": "An app key can no longer say yes to unlocking your door",- "summary": "A person saying yes is the one thing that stands between an agent and your front door, and we found a way around it. If you had ever issued an API key or connected an app to your three.ws account, that key could send an unlock request with the \"the owner already approved this\" flag set, and the lock opened, with nobody asked and nothing to click. We reproduced it against a real house and a real deadbolt before fixing it. Two things changed. Only a signed-in browser session can say yes now: a key or a connected app can ask to unlock, open or disarm, and is always answered with \"a person has to approve this\", exactly as the agent in your chat window already was. And the home permissions you approve when you connect an app are now actually enforced on every home request, so an app you gave read-only access to, or one you never gave home access at all, is refused instead of let through. Nothing legitimate changed: your agent still asks, you still approve, the door still opens.",- "link": "/smart-home",- "tags": [- "security",- "fix"- ]- },- {- "date": "2026-09-09",- "type": "update",- "title": "An embedded avatar no longer crashes the page it was removed from",- "summary": "Sites embed three.ws avatars with the <agent-3d> tag, and a single-page app can swap the whole page out while an avatar is still loading. When that happened the avatar was left holding a piece of the page that no longer existed, and its own tidy-up threw an error into the host site console instead of quietly reporting that loading had stopped. The avatar had already handled the failure correctly; the report of it was what broke. Embedders now get the agent:error event they listen for, and nothing else, whether or not the page moved on while the avatar was still loading.",- "link": "/docs/web-component",- "tags": [- "fix",- "sdk"- ]- },- {- "date": "2026-09-09",- "type": "update",- "title": "Arm your agent stops telling signed-out visitors to create an agent, and shows them the live conviction stream instead",- "summary": "A full audit of /oracle/arm found that anyone not signed in was shown “Create a 3D agent to arm it”, a dead end for a visitor with no account, and the rest of the page was hidden behind it. Signing in now gets its own state with sign-in and register buttons, and the live “clearing your bar” preview plus the 30-day win-rate readout stay on screen for everyone, because both run on public data. If the agents list itself fails to load, the page now says so with the status it got back and a Try again button, instead of pretending the account is empty. Four more fixes went with it: the arm switch and the two rule toggles can now be operated from the keyboard (they were focusable but did nothing on Enter or Space), coin artwork routes through the site image proxy so icons stop failing on IPFS gateways, the status line no longer flips back to “Disarmed” when the translation catalog lands after the page has rendered, and every form field now has a real label attached to it.",- "link": "/oracle/arm",- "tags": [- "fix",- "improvement"- ]- },- {- "date": "2026-09-09",- "type": "update",- "title": "Avatars and token art load again on sites that embed three.ws",- "summary": "Assets served from /cdn (avatars, thumbnails, generated models) are read from our object store with a signed request, and when that request is rejected the route falls back to the store's public address so the bytes still arrive. That fallback was a redirect, and a redirect throws away the cross-origin permission the route grants: browsers re-check permission at the redirect's destination, and the public address grants none. Measured from an outside site, every one of those reads failed outright, which took down model previews on pages that load art through the proxy and silently broke third-party embeds of three.ws avatars. The bytes are now served through three.ws itself rather than pointed at, so the permission, the file type and the download rules all survive the fallback, partial reads for video and audio keep working, and a file that genuinely is not there says so instead of reporting a server fault the page retries forever. It also keeps the fallback off the rate-limited public address, which is the whole reason the proxy exists.",- "link": null,- "tags": [- "fix"- ]- },- {- "date": "2026-09-09",- "type": "update",- "title": "Connecting a home: the page now says which state it is in, out loud",- "summary": "Two moments on /smart-home were doing the right thing silently. Typing an address that can never work (a 192.168 number, a .local name) has always put the reason under the field straight away, before anything is sent, but the message was carried by colour alone and a screen reader never read it out; it is now announced the moment it appears, and marked in a way that does not depend on seeing red. And a house that has stopped answering, is refusing the stored token, or has simply gone quiet is now labelled as such by the page itself rather than only in the sentence on its card. Nothing about what you see moves: the last known summary of your house stays exactly where it was, with how long ago it last answered, because a home that went quiet should never look like a home that was deleted.",- "link": "/smart-home",- "tags": [- "improvement",- "fix"- ]- },- {- "date": "2026-09-09",- "type": "update",- "title": "Disconnecting a home now stops the screens that were showing it",- "summary": "A kitchen tablet or wall display left open on a connected home kept saying \"Live\" and kept drawing that house after the home had been disconnected from another device. The connection was really gone, but the open page was never told, so it sat on a stale picture of somebody's house with a green Live badge on it. Disconnecting now closes the live connection and hands every screen still watching a final message that says what happened, and those screens stop asking for more. They keep showing the last state they saw, greyed and dated, with the reason and a Reconnect button, which is what every other kind of drop already did.",- "link": "/docs/home-scene",- "tags": [- "fix",- "security"- ]- },- {- "date": "2026-09-09",- "type": "update",- "title": "Endpoint Shopper fits a phone, and says something useful when a run finds nothing to buy",- "summary": "A full audit of /shopper, the page where you describe a task, set a budget, and let an agent shop the x402 Bazaar for endpoints that can answer it. Three things were wrong. A run that came back with no plan and no answer, which is what happens when nothing in the Bazaar matches the task or everything that does prices above your ceiling, painted an empty white column and stopped: no explanation, no way forward. That case now says no endpoints matched, quotes the task back, and offers the three things that change the outcome (rewrite the task, raise the budget, browse the Bazaar) plus a button to run again. Second, when the request could not reach us at all, the page showed the browser's own wording for a dropped connection, which names no cause and suggests no fix; it now says the agent service could not be reached and what to do about it. Third, on a phone narrower than 560px the four navigation links were squeezed into a strip about 76px wide and had to be dragged sideways a few pixels at a time to read; they now wrap onto their own full-width row. Alongside those: the results column announces itself to a screen reader while a run is in flight and when it settles, the Task label sits above its box as intended rather than beside it, and an internal ticket number that had been sitting in the page's headline label since launch is gone.",- "link": "/shopper",- "tags": [- "fix",- "improvement"- ]- },- {- "date": "2026-09-09",- "type": "update",- "title": "Ethereum wallets load on the portfolio page again, keylessly, and every token logo shows up",- "summary": "A full audit of /portfolio, where you paste any Solana or Ethereum address and get a live portfolio. The Ethereum half had gone dark: the single provider behind it ran out of monthly capacity, so every 0x address answered \"Data source unavailable\" with nothing behind it. Ethereum now has the same keyless floor Solana has always had. The ETH balance comes from the public Ethereum nodes the rest of the platform already uses, the token list with its prices and icons comes from a public block explorer, and the 24h move comes from the same market source the Solana side uses, with the coverage it achieved stated on the card. The Data panel names whichever lane actually answered, so you can see whether you are reading the paid provider or the free one. Token logos were another visible break: the page pointed your browser straight at an IPFS gateway, which the browser refuses on cross-origin grounds, so every holding fell back to a three-letter monogram and your address bar handed your IP to the gateway on every load. Logos now come through three.ws, sized for the row. Keyboard users could tab onto a slice of the allocation donut and get nothing: the tooltip was wired to an event handler property that browsers do not actually support, so it never fired once; the slice now names its asset and share, and Escape dismisses it. When a lookup fails, the page now shows the reason the server actually sent instead of a generic sentence, and clicking Retry after a rejected address no longer silently reloads the wallet you looked at before it. Underneath, the fallback that names Solana tokens when the primary metadata source is out of quota had never run at all, because of a missing import; token symbols and names come back on that path.",- "link": "/portfolio",- "tags": [- "fix",- "improvement"- ]}]}diff --git a/public/changelog.json b/public/changelog.jsonindex da9309be5..8fe0c5aad 100644--- a/public/changelog.json+++ b/public/changelog.json@@ -1,11 +1,61 @@{- "generated_at": "2026-09-30T08:30:10.207Z",+ "generated_at": "2026-09-30T08:35:45.854Z","generated_by": "scripts/build-page-index.mjs from data/pages.json + data/changelog.json","site": {"name": "three.ws","url": "https://three.ws"},"entries": [+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Docs · How we run and publish experiments",+ "summary": "The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.",+ "link": "/docs/experiments",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Docs · Trader duels",+ "summary": "How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.",+ "link": "/docs/trader-duels",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Experiment · Can autonomous agents trade new coins at a profit?",+ "summary": "1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.",+ "link": "/docs/experiments-autonomous-trading",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Experiment · How many new agents ever launch a coin?",+ "summary": "Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.",+ "link": "/docs/experiments-agent-launch-funnel",+ "tags": [+ "launch"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "launch",+ "title": "Experiments",+ "summary": "What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.",+ "link": "/experiments",+ "tags": [+ "launch"+ ]+ },{"date": "2026-09-30","type": "launch",@@ -67,6 +117,26 @@"feature"]},+ {+ "date": "2026-09-30",+ "type": "update",+ "title": "Experiment write-up: can autonomous agents trade new coins at a profit?",+ "summary": "The full record of our AI trading agents so far, measured from production: 1,296 real trades on Solana, 16.4% of them winners, and a net loss of 0.5402 SOL on 9.1444 SOL put to work. Only take-profit and take-initials exits made money, and the August retune made results worse, not better. The write-up says what that cost, what we got wrong, what we changed, and prints every query behind the numbers.",+ "link": "/docs/experiments-autonomous-trading",+ "tags": [+ "docs"+ ]+ },+ {+ "date": "2026-09-30",+ "type": "update",+ "title": "Experiment write-up: how many new agents ever launch a coin?",+ "summary": "Of the 788 agents people created in the 30 days to September 30, 2 went on to launch a coin, about three in four had no wallet, and the token plan built to bridge that gap has never been saved. The write-up publishes the numbers, what we got wrong about reading agent sign-ups as traction, what we already changed, and the queries so anyone can re-run them. It is the first entry on the new experiments page.",+ "link": "/docs/experiments-agent-launch-funnel",+ "tags": [+ "docs"+ ]+ },{"date": "2026-09-30","type": "update",@@ -149,6 +219,16 @@"improvement"]},+ {+ "date": "2026-09-30",+ "type": "update",+ "title": "Trader duels: call which trader wins, free to play",+ "summary": "A new Duels page pairs traders who sit next to each other on the board and asks one question: who books more realized profit over tomorrow, or over next week? Make your call before the window starts with free points from your daily allowance, then watch the race live on the duel page with both traders' records, trade rooms and profiles one tap away. Results come straight from the on-chain trade ledger, the same numbers their profiles show, and ties or empty windows refund every call. A correct call doubles your stake and adds XP to your quest level, and there is a monthly season board and four badges to earn. Points have no cash value and cannot be bought, sold, transferred or redeemed.",+ "link": "/duels",+ "tags": [+ "feature"+ ]+ },{"date": "2026-09-29","type": "launch",diff --git a/public/changelog.xml b/public/changelog.xmlindex f2d5994d0..db6385373 100644--- a/public/changelog.xml+++ b/public/changelog.xml@@ -6,6 +6,46 @@<atom:link href="https://three.ws/changelog.xml" rel="self" type="application/rss+xml"/><description>What's new on three.ws: features, improvements, and releases, newest first.</description><language>en</language>+ <item>+ <title>Docs · How we run and publish experiments</title>+ <link>https://three.ws/docs/experiments</link>+ <guid isPermaLink="false">2026-09-30:Docs · How we run and publish experiments</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>launch</category>+ <description>The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.</description>+ </item>+ <item>+ <title>Docs · Trader duels</title>+ <link>https://three.ws/docs/trader-duels</link>+ <guid isPermaLink="false">2026-09-30:Docs · Trader duels</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>launch</category>+ <description>How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.</description>+ </item>+ <item>+ <title>Experiment · Can autonomous agents trade new coins at a profit?</title>+ <link>https://three.ws/docs/experiments-autonomous-trading</link>+ <guid isPermaLink="false">2026-09-30:Experiment · Can autonomous agents trade new coins at a profit?</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>launch</category>+ <description>1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.</description>+ </item>+ <item>+ <title>Experiment · How many new agents ever launch a coin?</title>+ <link>https://three.ws/docs/experiments-agent-launch-funnel</link>+ <guid isPermaLink="false">2026-09-30:Experiment · How many new agents ever launch a coin?</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>launch</category>+ <description>Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.</description>+ </item>+ <item>+ <title>Experiments</title>+ <link>https://three.ws/experiments</link>+ <guid isPermaLink="false">2026-09-30:Experiments</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>launch</category>+ <description>What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.</description>+ </item><item><title>Platform Analytics</title><link>https://three.ws/analytics</link>@@ -54,6 +94,22 @@<category>feature</category><description>GET /api/platform/analytics returns agents, agents with a wallet, coins launched, 3D models generated, LLM tokens processed, x402 settlements and volume, marketplace sales and agent-to-agent hire volume, each with a daily series over 30 days, 90 days or all time and a plain sentence saying exactly which rows were counted. A figure that cannot be read comes back empty with a named reason instead of as zero. It powers the new analytics page and is open to anyone building on three.ws.</description></item>+ <item>+ <title>Experiment write-up: can autonomous agents trade new coins at a profit?</title>+ <link>https://three.ws/docs/experiments-autonomous-trading</link>+ <guid isPermaLink="false">2026-09-30:Experiment write-up: can autonomous agents trade new coins at a profit?</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>docs</category>+ <description>The full record of our AI trading agents so far, measured from production: 1,296 real trades on Solana, 16.4% of them winners, and a net loss of 0.5402 SOL on 9.1444 SOL put to work. Only take-profit and take-initials exits made money, and the August retune made results worse, not better. The write-up says what that cost, what we got wrong, what we changed, and prints every query behind the numbers.</description>+ </item>+ <item>+ <title>Experiment write-up: how many new agents ever launch a coin?</title>+ <link>https://three.ws/docs/experiments-agent-launch-funnel</link>+ <guid isPermaLink="false">2026-09-30:Experiment write-up: how many new agents ever launch a coin?</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>docs</category>+ <description>Of the 788 agents people created in the 30 days to September 30, 2 went on to launch a coin, about three in four had no wallet, and the token plan built to bridge that gap has never been saved. The write-up publishes the numbers, what we got wrong about reading agent sign-ups as traction, what we already changed, and the queries so anyone can re-run them. It is the first entry on the new experiments page.</description>+ </item><item><title>Photo to 3D keeps going when its main engines go down</title><link>https://three.ws/image-to-3d</link>@@ -118,6 +174,14 @@<category>improvement</category><description>The Oracle's conviction model is published data, and the @three-ws/oracle-model package is meant to score coins anywhere. Until now its download endpoint only answered pages on three.ws and a short list of partners, so a browser app on any other site could not load the model. It now answers every origin, the same as the rest of the Oracle's public reads, so any web page can fetch the model once and score coins locally with no key.</description></item>+ <item>+ <title>Trader duels: call which trader wins, free to play</title>+ <link>https://three.ws/duels</link>+ <guid isPermaLink="false">2026-09-30:Trader duels: call which trader wins, free to play</guid>+ <pubDate>Wed, 30 Sep 2026 12:00:00 GMT</pubDate>+ <category>feature</category>+ <description>A new Duels page pairs traders who sit next to each other on the board and asks one question: who books more realized profit over tomorrow, or over next week? Make your call before the window starts with free points from your daily allowance, then watch the race live on the duel page with both traders' records, trade rooms and profiles one tap away. Results come straight from the on-chain trade ledger, the same numbers their profiles show, and ties or empty windows refund every call. A correct call doubles your stake and adds XP to your quest level, and there is a monthly season board and four badges to earn. Points have no cash value and cannot be bought, sold, transferred or redeemed.</description>+ </item><item><title>Connect three.ws to Claude, ChatGPT, Cursor & VS Code</title><link>https://three.ws/connect</link>diff --git a/public/experiments.json b/public/experiments.jsonnew file mode 100644index 000000000..052fd2bf8--- /dev/null+++ b/public/experiments.json@@ -0,0 +1,50 @@+{+ "generated": true,+ "count": 2,+ "experiments": [+ {+ "slug": "autonomous-trading",+ "title": "Can autonomous agents trade new coins at a profit?",+ "question": "Given real SOL and a written strategy, do AI agents trading newly launched coins make money after fees and bad fills, and which strategies come closest?",+ "headline": "1,296 real trades, 16.4% winners, a net loss of 0.5402 SOL on 9.1444 SOL deployed. Only take-profit and take-initials exits made money.",+ "published": "2026-09-30",+ "spend_usd": 63.87,+ "status": "running",+ "doc": "experiments-autonomous-trading",+ "url": "/docs/experiments-autonomous-trading",+ "live_url": "/sniper/experiments",+ "related": [+ {+ "label": "The experiment spec",+ "href": "/docs/trading-experiment"+ },+ {+ "label": "The first 90 trades, in depth",+ "href": "/blog/autonomous-trading-experiment"+ }+ ]+ },+ {+ "slug": "agent-launch-funnel",+ "title": "How many new agents ever launch a coin?",+ "question": "Of the agents people create on three.ws, how many go on to launch a coin, and where does everyone else stop?",+ "headline": "2 of 788 agents created in 30 days launched a coin, about three in four had no wallet, and no token plan has ever been saved.",+ "published": "2026-09-30",+ "spend_usd": 0,+ "status": "concluded",+ "doc": "experiments-agent-launch-funnel",+ "url": "/docs/experiments-agent-launch-funnel",+ "live_url": null,+ "related": [+ {+ "label": "Live funnel numbers",+ "href": "/analytics"+ },+ {+ "label": "Launch a coin",+ "href": "/launch"+ }+ ]+ }+ ]+}diff --git a/public/features.json b/public/features.jsonindex 1c053b327..f2e95ba4b 100644--- a/public/features.json+++ b/public/features.json@@ -1,5 +1,5 @@{- "generated_at": "2026-09-30T08:30:10.181Z",+ "generated_at": "2026-09-30T08:35:45.828Z","generated_by": "scripts/build-page-index.mjs from data/pages.json","site": {"name": "three.ws",@@ -94,7 +94,8 @@"/awesome","/changelog","/blog",- "/status"+ "/status",+ "/experiments"]}]@@ -3006,6 +3007,16 @@"tags": [],"showcase": false},+ {+ "path": "/experiments",+ "title": "Experiments",+ "description": "What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.",+ "added": "2026-09-30",+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": false+ },{"path": "/analytics","title": "Platform Analytics",@@ -7357,6 +7368,36 @@"tags": [],"showcase": false},+ {+ "path": "/docs/experiments",+ "title": "Docs · How we run and publish experiments",+ "description": "The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.",+ "added": "2026-09-30",+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": false+ },+ {+ "path": "/docs/experiments-agent-launch-funnel",+ "title": "Experiment · How many new agents ever launch a coin?",+ "description": "Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.",+ "added": "2026-09-30",+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": false+ },+ {+ "path": "/docs/experiments-autonomous-trading",+ "title": "Experiment · Can autonomous agents trade new coins at a profit?",+ "description": "1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.",+ "added": "2026-09-30",+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": false+ },{"path": "/tutorials/getting-started","title": "Tutorial · Getting started (hosted)",@@ -7877,6 +7918,16 @@"tags": [],"showcase": false},+ {+ "path": "/docs/trader-duels",+ "title": "Docs · Trader duels",+ "description": "How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.",+ "added": "2026-09-30",+ "auth": null,+ "indexable": true,+ "tags": [],+ "showcase": false+ },{"path": "/docs/trading-quests","title": "Docs · Daily trading quests",diff --git a/public/footer.html b/public/footer.htmlindex e14bd9621..233ac20ab 100644--- a/public/footer.html+++ b/public/footer.html@@ -89,6 +89,7 @@<h3 class="h-footer-link-heading" data-i18n="footer.resources">Resources</h3><a href="/dashboard" data-i18n="footer.dashboard">Dashboard</a><a href="/analytics">Analytics</a>+ <a href="/experiments">Experiments</a><a href="/community" data-i18n="footer.community">Community</a><a href="https://community.ibm.com/community/user/groups/community-home?communitykey=e71510cc-d953-408f-9a1c-019f5c0a7016" target="_blank" rel="noopener" data-i18n="footer.ibm_user_group">IBM user group</a><a href="/tutorials" data-i18n="footer.tutorials">Tutorials</a>diff --git a/public/llms-full.txt b/public/llms-full.txtindex 5f257f8f4..54e503cc9 100644--- a/public/llms-full.txt+++ b/public/llms-full.txt@@ -1789,6 +1789,12 @@ URL: https://three.ws/agent-economy-volumeLive dashboard of total agent-to-agent economy volume: real USDC settled between AI agents hiring each other for paid skills over x402, with top earners, top spenders, and a 90-day volume chart.+### Experiments++URL: https://three.ws/experiments++What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.+### Platform AnalyticsURL: https://three.ws/analytics@@ -4421,6 +4427,24 @@ URL: https://three.ws/docs/trading-experimentThe full risk policy behind the autonomous fleet: entry band, position sizing, taking initials at 2x, the moon-bag floor, trailing and hard stops, and how to read the decision journal.+### Docs · How we run and publish experiments++URL: https://three.ws/docs/experiments++The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.++### Experiment · How many new agents ever launch a coin?++URL: https://three.ws/docs/experiments-agent-launch-funnel++Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.++### Experiment · Can autonomous agents trade new coins at a profit?++URL: https://three.ws/docs/experiments-autonomous-trading++1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.+### Tutorial · Getting started (hosted)URL: https://three.ws/tutorials/getting-started@@ -4733,6 +4757,12 @@ URL: https://three.ws/docs/syndicatesHow copy-trading syndicates work: founding around copyable leaders, joining with your own wallet and caps, how the group curve and rival standing are computed from real copies, and the full API.+### Docs · Trader duels++URL: https://three.ws/docs/trader-duels++How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.+### Docs · Daily trading questsURL: https://three.ws/docs/trading-questsdiff --git a/public/llms.txt b/public/llms.txtindex 3c08424da..f2569b0c4 100644--- a/public/llms.txt+++ b/public/llms.txt@@ -92,9 +92,10 @@ Machine-readable entry points. Fetch these instead of scraping HTML.- [Changelog](https://three.ws/changelog): What is new on three.ws: features, improvements, and releases, newest first.- [Blog](https://three.ws/blog): Editorial index of three.ws launches and long-form posts.- [Status](https://three.ws/status): Live operational status and 90-day uptime history for the platform, API, and x402 surfaces: probed every 5 minutes.+- [Experiments](https://three.ws/experiments): What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.## Optional-- [llms-full.txt](https://three.ws/llms-full.txt): Every public page (918), grouped by section, with full descriptions.+- [llms-full.txt](https://three.ws/llms-full.txt): Every public page (923), grouped by section, with full descriptions.- [Sitemap](https://three.ws/sitemap): The same index for people, with search.- [sitemap.xml](https://three.ws/sitemap.xml): XML sitemap for crawlers, including every agent and avatar page.diff --git a/public/nav-data.js b/public/nav-data.jsindex 95be329da..df5f7a0c0 100644--- a/public/nav-data.js+++ b/public/nav-data.js@@ -336,6 +336,12 @@ export const NAV_GROUPS = [href: '/three-token',desc: 'Live price, bonding-curve chart, streaming trades & one-click buy',},+ {+ title: 'Experiments',+ href: '/experiments',+ tier: 'advanced',+ desc: 'What we tried, what it cost, what the numbers said and what we got wrong, with the queries behind every number',+ },{title: 'Platform Analytics',href: '/analytics',diff --git a/public/sitemap/index.html b/public/sitemap/index.htmlindex c484a2984..cc6717b6a 100644--- a/public/sitemap/index.html+++ b/public/sitemap/index.html@@ -3,16 +3,16 @@<head><meta charset="utf-8" /><meta name="viewport" content="width=device-width,initial-scale=1" />-<title data-i18n="sitemap_index.meta_title">Sitemap · three.ws</title>-<meta name="description" content="Human-readable index of every page on three.ws." data-i18n-attr="content:sitemap_index.meta_description"/>+<title>Sitemap · three.ws</title>+<meta name="description" content="Human-readable index of every page on three.ws." /><meta name="robots" content="index, follow, max-image-preview:large, max-snippet:-1" /><link rel="canonical" href="https://three.ws/sitemap" /><meta property="og:type" content="website" /><meta property="og:site_name" content="three.ws" /><meta property="og:locale" content="en_US" /><meta property="og:url" content="https://three.ws/sitemap" />-<meta property="og:title" content="Sitemap · three.ws" data-i18n-attr="content:sitemap_index.meta_og_title"/>-<meta property="og:description" content="Human-readable index of every page on three.ws." data-i18n-attr="content:sitemap_index.meta_og_description"/>+<meta property="og:title" content="Sitemap · three.ws" />+<meta property="og:description" content="Human-readable index of every page on three.ws." /><meta property="og:image" content="https://three.ws/api/page-og?s=main&t=Sitemap&d=Human-readable+index+of+every+page+on+three.ws.&p=%2Fsitemap" /><meta property="og:image:width" content="1200" /><meta property="og:image:height" content="630" />@@ -20,8 +20,8 @@<meta property="og:image:alt" content="Sitemap - three.ws" /><meta name="twitter:card" content="summary_large_image" /><meta name="twitter:site" content="@trythreews" />-<meta name="twitter:title" content="Sitemap · three.ws" data-i18n-attr="content:sitemap_index.meta_twitter_title"/>-<meta name="twitter:description" content="Human-readable index of every page on three.ws." data-i18n-attr="content:sitemap_index.meta_twitter_description"/>+<meta name="twitter:title" content="Sitemap · three.ws" />+<meta name="twitter:description" content="Human-readable index of every page on three.ws." /><meta name="twitter:image" content="https://three.ws/api/page-og?s=main&t=Sitemap&d=Human-readable+index+of+every+page+on+three.ws.&p=%2Fsitemap" /><script type="application/ld+json">{"@context":"https://schema.org","@graph":[{"@type":"WebPage","name":"Sitemap","description":"Human-readable index of every page on three.ws.","url":"https://three.ws/sitemap","isPartOf":{"@type":"WebSite","name":"three.ws","url":"https://three.ws"},"primaryImageOfPage":"https://three.ws/api/page-og?s=main&t=Sitemap&d=Human-readable+index+of+every+page+on+three.ws.&p=%2Fsitemap"}]}</script><link rel="alternate" type="application/xml" title="XML sitemap" href="/sitemap.xml" />@@ -95,26 +95,26 @@<header><div id="nav-container"></div></header><main class="sm-wrap"><div class="sm-hero">- <h1 data-i18n="sitemap_index.sitemap">Sitemap</h1>- <p><span>968</span> <span data-i18n="sitemap_index.pages_grouped">pages on three.ws, grouped by purpose. Looking for the machine-readable versions?</span></p>+ <h1>Sitemap</h1>+ <p><span>973</span> <span data-i18n="sitemap_index.pages_grouped">pages on three.ws, grouped by purpose. Looking for the machine-readable versions?</span></p><div class="sm-formats">- <a href="/sitemap.xml" data-i18n-html="sitemap_index.for_search_engines"><code>sitemap.xml</code> · for search engines</a>- <a href="/llms.txt" data-i18n-html="sitemap_index.for_ai_agents"><code>llms.txt</code> · for AI agents</a>- <a href="/llms-full.txt" data-i18n-html="sitemap_index.long_form"><code>llms-full.txt</code> · long form</a>- <a href="/api/features.json" data-i18n-html="sitemap_index.machine_manifest"><code>features.json</code> · machine manifest</a>- <a href="/changelog" data-i18n-html="sitemap_index.what_s_new"><code>changelog</code> · what's new</a>- <a href="/openapi.json" data-i18n-html="sitemap_index.http_api"><code>openapi.json</code> · HTTP API</a>+ <a href="/sitemap.xml"><code>sitemap.xml</code> · for search engines</a>+ <a href="/llms.txt"><code>llms.txt</code> · for AI agents</a>+ <a href="/llms-full.txt"><code>llms-full.txt</code> · long form</a>+ <a href="/api/features.json"><code>features.json</code> · machine manifest</a>+ <a href="/changelog"><code>changelog</code> · what's new</a>+ <a href="/openapi.json"><code>openapi.json</code> · HTTP API</a></div></div><div class="sm-filter"><svg viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2" stroke-linecap="round" aria-hidden="true"><circle cx="11" cy="11" r="7"/><path d="m21 21-4.3-4.3"/></svg>- <input id="sm-filter-input" type="search" placeholder="Filter 968 pages by name, path, or description…" aria-label="Filter pages" autocomplete="off" spellcheck="false" />+ <input id="sm-filter-input" type="search" placeholder="Filter 973 pages by name, path, or description…" aria-label="Filter pages" autocomplete="off" spellcheck="false" /><kbd id="sm-filter-kbd" aria-hidden="true">/</kbd></div><p class="sm-filter-count" role="status" aria-live="polite"><span id="sm-filter-count"></span><span id="sm-filter-hint" aria-hidden="true"></span></p><section class="sm-new" id="sm-new" aria-labelledby="sm-new-title">- <h2 id="sm-new-title" data-i18n-html="sitemap_index.newestfull_changelog">Newest<a class="sm-new-more" href="/changelog">full changelog →</a></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.the_9_most_recently_launched_pages_straight">The 9 most recently launched pages, straight to the live surface.</p>+ <h2 id="sm-new-title">Newest<a class="sm-new-more" href="/changelog">full changelog →</a></h2>+ <p class="sm-section-desc">The 9 most recently launched pages, straight to the live surface.</p><ul class="sm-list"><li><a href="/duels">@@ -124,6 +124,14 @@<span class="sm-desc" data-i18n="sitemap_index.d_duels">Call which of two pump.fun trader agents books more realized profit over a fixed UTC window. Free to play with a daily points allowance; points have no cash value. Resolved from the on-chain trade ledger, with a season leaderboard and badges.</span></a></li>+ <li>+ <a href="/experiments">+ <span class="sm-new-date" translate="no">Sep 30, 2026</span>+ <span class="sm-title"><span data-i18n="sitemap_index.t_experiments">Experiments</span></span>+ <span class="sm-path" translate="no">/experiments</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_experiments">What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.</span>+ </a>+ </li><li><a href="/analytics"><span class="sm-new-date" translate="no">Sep 30, 2026</span>@@ -141,72 +149,64 @@</a></li><li>- <a href="/trade-rooms">- <span class="sm-new-date" translate="no">Sep 29, 2026</span>- <span class="sm-title"><span data-i18n="sitemap_index.t_trade_rooms">Live Trade Rooms: watch pump.fun traders trade live in 3D</span></span>- <span class="sm-path" translate="no">/trade-rooms</span>- <span class="sm-desc" data-i18n="sitemap_index.d_trade_rooms">Step into a trader's 3D room and watch their real pump.fun buys and sells land as they fill: the avatar reacts, every trade carries its receipt, P&L and Solscan tx, and one tap forks the trade or ghost-copies the trader. Quiet rooms show the latest real trades and which rooms are live now.</span>- </a>- </li>- <li>- <a href="/copy-coach">- <span class="sm-new-date" translate="no">Sep 29, 2026</span>- <span class="sm-title"><span data-i18n="sitemap_index.t_copy_coach">Copy Coach: your first copy trade, the careful way</span></span>- <span class="sm-path" translate="no">/copy-coach</span>- <span class="sm-desc" data-i18n="sitemap_index.d_copy_coach">A guided first copy trade for someone who has never copied one. See a real verified win next to the full record behind it, ghost-copy a leader with fake money, then make one tiny, server-capped copy you sign yourself. The coach answers from the real numbers on the page and never promises returns.</span>+ <a href="/docs/experiments">+ <span class="sm-new-date" translate="no">Sep 30, 2026</span>+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_experiments">Docs · How we run and publish experiments</span></span>+ <span class="sm-path" translate="no">/docs/experiments</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_experiments">The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.</span></a></li><li>- <a href="/syndicates">- <span class="sm-new-date" translate="no">Sep 29, 2026</span>- <span class="sm-title"><span data-i18n="sitemap_index.t_syndicates">Syndicates: copy-trade as a team</span></span>- <span class="sm-path" translate="no">/syndicates</span>- <span class="sm-desc" data-i18n="sitemap_index.d_syndicates">Named teams of copiers who follow the same verified pump.fun trader agents together. Each member copies from their own wallet under their own caps; the team shares a flag, a live equity curve from members' real copies, and a rival on the board. No pooled funds, no custody.</span>+ <a href="/docs/experiments-agent-launch-funnel">+ <span class="sm-new-date" translate="no">Sep 30, 2026</span>+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_experiments_agent_launch_funnel">Experiment · How many new agents ever launch a coin?</span></span>+ <span class="sm-path" translate="no">/docs/experiments-agent-launch-funnel</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_experiments_agent_launch_funnel">Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.</span></a></li><li>- <a href="/quests">- <span class="sm-new-date" translate="no">Sep 29, 2026</span>- <span class="sm-title"><span data-i18n="sitemap_index.t_quests">Daily Trading Quests</span></span>- <span class="sm-path" translate="no">/quests</span>- <span class="sm-desc" data-i18n="sitemap_index.d_quests">Five trading quests that reset every UTC midnight: make three trades, fork a verified trade, ghost-copy a new agent, act on a copy intent, and let your agent close a winner. Scored from on-chain proof, rewarded with XP, levels and badges.</span>+ <a href="/docs/experiments-autonomous-trading">+ <span class="sm-new-date" translate="no">Sep 30, 2026</span>+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_experiments_autonomous_trading">Experiment · Can autonomous agents trade new coins at a profit?</span></span>+ <span class="sm-path" translate="no">/docs/experiments-autonomous-trading</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_experiments_autonomous_trading">1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.</span></a></li><li>- <a href="/docs/cli">- <span class="sm-new-date" translate="no">Sep 29, 2026</span>- <span class="sm-title"><span data-i18n="sitemap_index.t_docs_cli">Docs · three-ws CLI</span></span>- <span class="sm-path" translate="no">/docs/cli</span>- <span class="sm-desc" data-i18n="sitemap_index.d_docs_cli">Connect Claude Code, Claude Desktop, Cursor, Windsurf, VS Code, Codex, Gemini CLI or Hermes to three.ws in one command: npx three-ws setup signs you in, writes every client's MCP config and verifies each server with a live tools/list.</span>+ <a href="/docs/trader-duels">+ <span class="sm-new-date" translate="no">Sep 30, 2026</span>+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_trader_duels">Docs · Trader duels</span></span>+ <span class="sm-path" translate="no">/docs/trader-duels</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_trader_duels">How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.</span></a></li><li>- <a href="/connect">+ <a href="/trade-rooms"><span class="sm-new-date" translate="no">Sep 29, 2026</span>- <span class="sm-title"><span data-i18n="sitemap_index.t_connect">Connect three.ws to Claude, ChatGPT, Cursor & VS Code</span></span>- <span class="sm-path" translate="no">/connect</span>- <span class="sm-desc" data-i18n="sitemap_index.d_connect">Add three.ws to your AI client in two clicks: one connector URL gives Claude, ChatGPT, Cursor, VS Code or Claude Code the three.ws MCP tools for 3D generation, agents, wallets and x402 payments. Sign in once in your browser, with no API key to copy.</span>+ <span class="sm-title"><span data-i18n="sitemap_index.t_trade_rooms">Live Trade Rooms: watch pump.fun traders trade live in 3D</span></span>+ <span class="sm-path" translate="no">/trade-rooms</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_trade_rooms">Step into a trader's 3D room and watch their real pump.fun buys and sells land as they fill: the avatar reacts, every trade carries its receipt, P&L and Solscan tx, and one tap forks the trade or ghost-copies the trader. Quiet rooms show the latest real trades and which rooms are live now.</span></a></li></ul></section><nav class="sm-toc" aria-label="Sections">- <a href="#main" data-i18n="sitemap_index.main">Main</a>- <a href="#build" data-i18n="sitemap_index.build">Build</a>- <a href="#labs" data-i18n="sitemap_index.labs">Labs</a>- <a href="#crypto" data-i18n="sitemap_index.crypto">Crypto</a>- <a href="#agent-tools" data-i18n="sitemap_index.agent_tools">Agent Tools</a>- <a href="#account" data-i18n="sitemap_index.account">Account</a>- <a href="#learn" data-i18n="sitemap_index.learn">Learn</a>- <a href="#blog" data-i18n="sitemap_index.blog">Blog</a>- <a href="#legal" data-i18n="sitemap_index.legal">Legal</a>- <a href="#machine" data-i18n="sitemap_index.machine_readable">Machine-readable</a>- <a href="#news" data-i18n="sitemap_index.news">News</a>+ <a href="#main">Main</a>+ <a href="#build">Build</a>+ <a href="#labs">Labs</a>+ <a href="#crypto">Crypto</a>+ <a href="#agent-tools">Agent Tools</a>+ <a href="#account">Account</a>+ <a href="#learn">Learn</a>+ <a href="#blog">Blog</a>+ <a href="#legal">Legal</a>+ <a href="#machine">Machine-readable</a>+ <a href="#news">News</a></nav><section class="sm-section" id="main"><header><h2><span data-i18n="sitemap_index.section_main">Main</span><span class="sm-count" data-count>62</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.entry_points_into_the_product">Entry points into the product.</p>+ <p class="sm-section-desc">Entry points into the product.</p></header><ul class="sm-list"><li>@@ -649,7 +649,7 @@<section class="sm-section" id="build"><header><h2><span data-i18n="sitemap_index.section_build">Build</span><span class="sm-count" data-count>80</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.create_and_customize_an_agent">Create and customize an agent.</p>+ <p class="sm-section-desc">Create and customize an agent.</p></header><ul class="sm-list"><li>@@ -1218,7 +1218,7 @@<section class="sm-section" id="labs"><header><h2><span data-i18n="sitemap_index.section_labs">Labs</span><span class="sm-count" data-count>25</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.experimental_features_actively_shipping_expe">Experimental features actively shipping. Expect fast iteration.</p>+ <p class="sm-section-desc">Experimental features actively shipping. Expect fast iteration.</p></header><ul class="sm-list"><li>@@ -1401,8 +1401,8 @@<section class="sm-section" id="crypto"><header>- <h2><span data-i18n="sitemap_index.section_crypto">Crypto</span><span class="sm-count" data-count>153</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.on_chain_features_pay_per_call_token">On-chain features: pay-per-call, token launches, vanity wallets.</p>+ <h2><span data-i18n="sitemap_index.section_crypto">Crypto</span><span class="sm-count" data-count>154</span></h2>+ <p class="sm-section-desc">On-chain features: pay-per-call, token launches, vanity wallets.</p></header><ul class="sm-list"><li>@@ -2252,6 +2252,13 @@<span class="sm-desc" data-i18n="sitemap_index.d_agent_economy_volume">Live dashboard of total agent-to-agent economy volume: real USDC settled between AI agents hiring each other for paid skills over x402, with top earners, top spenders, and a 90-day volume chart.</span></a></li>+ <li>+ <a href="/experiments">+ <span class="sm-title"><span data-i18n="sitemap_index.t_experiments">Experiments</span></span>+ <span class="sm-path" translate="no">/experiments</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_experiments">What three.ws tried, what it cost, what the numbers said and what we got wrong. Every write-up is measured from production data and ends with the queries that produced its numbers, filterable by running, concluded or killed.</span>+ </a>+ </li><li><a href="/analytics"><span class="sm-title"><span data-i18n="sitemap_index.t_analytics">Platform Analytics</span></span>@@ -2482,7 +2489,7 @@<section class="sm-section" id="agent-tools"><header><h2><span data-i18n="sitemap_index.section_agent-tools">Agent Tools</span><span class="sm-count" data-count>10</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.manage_talk_to_and_deploy_individual_agents">Manage, talk to, and deploy individual agents.</p>+ <p class="sm-section-desc">Manage, talk to, and deploy individual agents.</p></header><ul class="sm-list"><li>@@ -2561,7 +2568,7 @@<section class="sm-section" id="account"><header><h2><span data-i18n="sitemap_index.section_account">Account</span><span class="sm-count" data-count>14</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.sign_in_dashboard_billing">Sign-in, dashboard, billing.</p>+ <p class="sm-section-desc">Sign-in, dashboard, billing.</p></header><ul class="sm-list"><li>@@ -2667,8 +2674,8 @@<section class="sm-section" id="learn"><header>- <h2><span data-i18n="sitemap_index.section_learn">Learn</span><span class="sm-count" data-count>447</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.docs_tutorials_and_reference">Docs, tutorials, and reference.</p>+ <h2><span data-i18n="sitemap_index.section_learn">Learn</span><span class="sm-count" data-count>451</span></h2>+ <p class="sm-section-desc">Docs, tutorials, and reference.</p></header><ul class="sm-list"><li>@@ -5310,6 +5317,27 @@<span class="sm-desc" data-i18n="sitemap_index.d_docs_trading_experiment">The full risk policy behind the autonomous fleet: entry band, position sizing, taking initials at 2x, the moon-bag floor, trailing and hard stops, and how to read the decision journal.</span></a></li>+ <li>+ <a href="/docs/experiments">+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_experiments">Docs · How we run and publish experiments</span></span>+ <span class="sm-path" translate="no">/docs/experiments</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_experiments">The six-section write-up template every three.ws experiment follows (question, method, spend, result, what we got wrong, what we changed), the data/experiments.json entry format, and the build check that enforces both.</span>+ </a>+ </li>+ <li>+ <a href="/docs/experiments-agent-launch-funnel">+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_experiments_agent_launch_funnel">Experiment · How many new agents ever launch a coin?</span></span>+ <span class="sm-path" translate="no">/docs/experiments-agent-launch-funnel</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_experiments_agent_launch_funnel">Measured from production: of 788 agents created in the 30 days to 2026-09-30, 2 launched a coin, about three in four had no wallet, and no token plan has ever been saved. What we got wrong and what we changed, with every query.</span>+ </a>+ </li>+ <li>+ <a href="/docs/experiments-autonomous-trading">+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_experiments_autonomous_trading">Experiment · Can autonomous agents trade new coins at a profit?</span></span>+ <span class="sm-path" translate="no">/docs/experiments-autonomous-trading</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_experiments_autonomous_trading">1,296 real trades by AI agents on Solana: 16.4% winners and a net loss of 0.5402 SOL on 9.1444 SOL deployed. Which exits made money, what the retune got wrong, and every query behind the numbers.</span>+ </a>+ </li><li><a href="/tutorials/getting-started"><span class="sm-title"><span data-i18n="sitemap_index.t_tutorials_getting_started">Tutorial · Getting started (hosted)</span></span>@@ -5674,6 +5702,13 @@<span class="sm-desc" data-i18n="sitemap_index.d_docs_syndicates">How copy-trading syndicates work: founding around copyable leaders, joining with your own wallet and caps, how the group curve and rival standing are computed from real copies, and the full API.</span></a></li>+ <li>+ <a href="/docs/trader-duels">+ <span class="sm-title"><span data-i18n="sitemap_index.t_docs_trader_duels">Docs · Trader duels</span></span>+ <span class="sm-path" translate="no">/docs/trader-duels</span>+ <span class="sm-desc" data-i18n="sitemap_index.d_docs_trader_duels">How trader duels are generated from the rivalry engine, how calls lock, how a duel resolves from the realized P&L ledger (ties, empty windows, idle and private traders), points, XP, seasons, badges, the anti-abuse rules and the full API. Free to play; points have no cash value.</span>+ </a>+ </li><li><a href="/docs/trading-quests"><span class="sm-title"><span data-i18n="sitemap_index.t_docs_trading_quests">Docs · Daily trading quests</span></span>@@ -5806,7 +5841,7 @@<section class="sm-section" id="blog"><header><h2><span data-i18n="sitemap_index.section_blog">Blog</span><span class="sm-count" data-count>39</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.long_form_announcements_launches_and_writing">Long-form announcements, launches, and writing from three.ws.</p>+ <p class="sm-section-desc">Long-form announcements, launches, and writing from three.ws.</p></header><ul class="sm-list"><li>@@ -6160,7 +6195,7 @@<section class="sm-section" id="machine"><header><h2><span data-i18n="sitemap_index.section_machine">Machine-readable</span><span class="sm-count" data-count>17</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.indexes_and_well_known_files_for_crawlers">Indexes and well-known files for crawlers and AI agents.</p>+ <p class="sm-section-desc">Indexes and well-known files for crawlers and AI agents.</p></header><ul class="sm-list"><li>@@ -6288,7 +6323,7 @@<section class="sm-section" id="news"><header><h2><span data-i18n="sitemap_index.section_news">News</span><span class="sm-count" data-count>112</span></h2>- <p class="sm-section-desc" data-i18n="sitemap_index.product_launches_integrations_and_announceme">Product launches, integrations, and announcements from three.ws.</p>+ <p class="sm-section-desc">Product launches, integrations, and announcements from three.ws.</p></header><ul class="sm-list"><li>@@ -7079,10 +7114,10 @@</section><div class="sm-empty" id="sm-empty" hidden><p><span data-i18n="sitemap_index.no_pages_match">No pages match</span> “<span id="sm-empty-q"></span>”.</p>- <p class="sm-empty-hint" data-i18n="sitemap_index.try_a_shorter_word_or_search_everything">Try a shorter word, or search everything: global search covers agents, coins, and skills too.</p>+ <p class="sm-empty-hint">Try a shorter word, or search everything: global search covers agents, coins, and skills too.</p><div class="sm-empty-actions">- <button type="button" id="sm-empty-clear" data-i18n="sitemap_index.clear_filter">Clear filter</button>- <a id="sm-empty-search" href="/search" data-i18n="sitemap_index.search_all_of_three_ws">Search all of three.ws</a>+ <button type="button" id="sm-empty-clear">Clear filter</button>+ <a id="sm-empty-search" href="/search">Search all of three.ws</a></div></div></main>diff --git a/scripts/build-page-index.mjs b/scripts/build-page-index.mjsindex e22cd4772..a7bad1d9a 100644--- a/scripts/build-page-index.mjs+++ b/scripts/build-page-index.mjs@@ -27,6 +27,7 @@ import { readFileSync, writeFileSync, mkdirSync, existsSync } from 'node:fs';import { execFileSync } from 'node:child_process';import { resolve, dirname } from 'node:path';import { fileURLToPath } from 'node:url';+import { loadExperiments } from './lib/experiments-index.mjs';const here = dirname(fileURLToPath(import.meta.url));const root = resolve(here, '..');@@ -98,6 +99,11 @@ for (const [index, e] of curatedEntries.entries()) {identicalEntries.set(fingerprint, index);}+// The public experiment log (/experiments). Validated as hard as the changelog:+// a missing field, an unknown status or a write-up without the six template+// sections fails the build. See scripts/lib/experiments-index.mjs.+const experimentsIndex = loadExperiments(root);+// Splice in the "News" section so news entries appear in the sitemap,// llms.txt, and the human-readable /sitemap page.//@@ -874,6 +880,7 @@ const outputs = [{ file: resolve(publicDir, 'changelog.json'), content: buildChangelogJson() },{ file: resolve(publicDir, 'changelog-recent.json'), content: buildChangelogRecentJson() },{ file: resolve(publicDir, 'changelog.xml'), content: buildChangelogRss() },+ { file: resolve(publicDir, 'experiments.json'), content: experimentsIndex.json },];let wrote = 0;diff --git a/scripts/lib/experiments-index.mjs b/scripts/lib/experiments-index.mjsnew file mode 100644index 000000000..e28d85517--- /dev/null+++ b/scripts/lib/experiments-index.mjs@@ -0,0 +1,125 @@+// The public experiment log: validation and the published index.+//+// data/experiments.json lists every experiment three.ws has run in public; each+// one points at a write-up in docs/<doc>.md that follows the template in+// docs/experiments.md. scripts/build-page-index.mjs calls validateExperiments()+// on every `npm run build:pages` and fails the build on the first problem, then+// writes buildExperimentsJson() to public/experiments.json for the /experiments+// page. A bad entry therefore never reaches the site.++import { existsSync, readFileSync } from 'node:fs';+import { resolve } from 'node:path';++export const EXPERIMENT_STATUSES = Object.freeze(['running', 'concluded', 'killed']);++// The six template sections, in the order every write-up must carry them.+export const WRITEUP_SECTIONS = Object.freeze([+ 'The question',+ 'Method',+ 'Spend',+ 'Result',+ 'What we got wrong',+ 'What we changed because of it',+]);++const REQUIRED = ['slug', 'title', 'question', 'published', 'spend_usd', 'status', 'doc'];+const SLUG_RE = /^[a-z0-9]+(?:-[a-z0-9]+)*$/;+const DATE_RE = /^\d{4}-\d{2}-\d{2}$/;++/** The `## ` headings of a markdown document, in order. */+export function markdownHeadings(markdown) {+ return String(markdown)+ .split('\n')+ .filter((line) => /^## /.test(line))+ .map((line) => line.slice(3).trim());+}++/**+ * Throw on the first invalid entry. `readDoc(slug)` returns the write-up's+ * markdown, or null when it does not exist.+ *+ * @param {unknown} data parsed data/experiments.json+ * @param {{ readDoc: (doc: string) => string | null }} io+ * @returns {Array<object>} the validated entries+ */+export function validateExperiments(data, { readDoc }) {+ const entries = data && Array.isArray(data.experiments) ? data.experiments : null;+ if (!entries) throw new Error('data/experiments.json: expected an "experiments" array');+ const seen = new Set();+ entries.forEach((e, i) => {+ const ctx = `data/experiments.json entry ${i} (${e?.slug || 'no slug'})`;+ for (const field of REQUIRED) {+ if (e?.[field] === undefined || e[field] === null || e[field] === '') {+ throw new Error(`${ctx}: missing required field "${field}"`);+ }+ }+ if (!SLUG_RE.test(e.slug)) throw new Error(`${ctx}: slug must be lowercase kebab-case`);+ if (seen.has(e.slug)) throw new Error(`${ctx}: duplicate slug "${e.slug}"`);+ seen.add(e.slug);+ if (!DATE_RE.test(e.published) || Number.isNaN(Date.parse(`${e.published}T00:00:00Z`))) {+ throw new Error(`${ctx}: published must be a real YYYY-MM-DD date`);+ }+ if (typeof e.spend_usd !== 'number' || !Number.isFinite(e.spend_usd) || e.spend_usd < 0) {+ throw new Error(`${ctx}: spend_usd must be a number of dollars, 0 or more`);+ }+ if (!EXPERIMENT_STATUSES.includes(e.status)) {+ throw new Error(`${ctx}: status must be one of ${EXPERIMENT_STATUSES.join(', ')}`);+ }+ if (e.doc !== `experiments-${e.slug}`) {+ throw new Error(`${ctx}: doc must be "experiments-${e.slug}" (the write-up lives at docs/experiments-${e.slug}.md)`);+ }+ const markdown = readDoc(e.doc);+ if (markdown == null) throw new Error(`${ctx}: write-up docs/${e.doc}.md does not exist`);+ const headings = markdownHeadings(markdown);+ let from = 0;+ for (const section of WRITEUP_SECTIONS) {+ const at = headings.indexOf(section, from);+ if (at === -1) throw new Error(`${ctx}: docs/${e.doc}.md is missing the "## ${section}" section, or has it out of order`);+ from = at + 1;+ }+ if (e.live_url !== undefined && !(typeof e.live_url === 'string' && e.live_url.startsWith('/'))) {+ throw new Error(`${ctx}: live_url must be a site path starting with "/"`);+ }+ if (e.related !== undefined) {+ if (!Array.isArray(e.related)) throw new Error(`${ctx}: related must be an array of { label, href }`);+ for (const r of e.related) {+ if (!r?.label || typeof r.href !== 'string' || !r.href.startsWith('/')) {+ throw new Error(`${ctx}: every related link needs a label and a site path href`);+ }+ }+ }+ });+ return entries;+}++/** The public index the /experiments page reads, newest first. */+export function buildExperimentsJson(entries) {+ const experiments = [...entries]+ .sort((a, b) => (a.published < b.published ? 1 : a.published > b.published ? -1 : a.title.localeCompare(b.title)))+ .map((e) => ({+ slug: e.slug,+ title: e.title,+ question: e.question,+ headline: e.headline || null,+ published: e.published,+ spend_usd: e.spend_usd,+ status: e.status,+ doc: e.doc,+ url: `/docs/${e.doc}`,+ live_url: e.live_url || null,+ related: e.related || [],+ }));+ return `${JSON.stringify({ generated: true, count: experiments.length, experiments }, null, '\t')}\n`;+}++/** Read, validate and render in one call, for scripts/build-page-index.mjs. */+export function loadExperiments(root) {+ const data = JSON.parse(readFileSync(resolve(root, 'data/experiments.json'), 'utf8'));+ const entries = validateExperiments(data, {+ readDoc: (doc) => {+ const file = resolve(root, 'docs', `${doc}.md`);+ return existsSync(file) ? readFileSync(file, 'utf8') : null;+ },+ });+ return { entries, json: buildExperimentsJson(entries) };+}diff --git a/src/experiments-index.js b/src/experiments-index.jsnew file mode 100644index 000000000..2c2e6ac30--- /dev/null+++ b/src/experiments-index.js@@ -0,0 +1,133 @@+/**+ * /experiments: the public experiment log. Reads /experiments.json, which+ * `npm run build:pages` validates and publishes from data/experiments.json, and+ * renders one card per experiment with a URL-synced status filter.+ */++const INDEX_URL = '/experiments.json';+const STATUSES = ['all', 'running', 'concluded', 'killed'];+const STATUS_LABEL = { running: 'Running', concluded: 'Concluded', killed: 'Killed' };++const state = { status: 'all', experiments: null };++const listEl = document.getElementById('ex-list');+const filterEl = document.getElementById('ex-filter');++function escapeHtml(s) {+ return String(s ?? '').replace(/[&<>"']/g, (c) => ({ '&': '&', '<': '<', '>': '>', '"': '"', "'": ''' })[c]);+}++function fmtDate(day) {+ const t = Date.parse(`${day}T00:00:00Z`);+ if (!Number.isFinite(t)) return day;+ return new Date(t).toLocaleDateString(undefined, { year: 'numeric', month: 'short', day: 'numeric', timeZone: 'UTC' });+}++function fmtSpend(usd) {+ const n = Number(usd);+ if (!Number.isFinite(n)) return 'not stated';+ if (n === 0) return '$0';+ return n.toLocaleString(undefined, { style: 'currency', currency: 'USD', minimumFractionDigits: 2, maximumFractionDigits: 2 });+}++function readUrl() {+ const s = new URLSearchParams(location.search).get('status');+ state.status = STATUSES.includes(s) ? s : 'all';+}++function writeUrl() {+ const q = new URLSearchParams(location.search);+ if (state.status === 'all') q.delete('status');+ else q.set('status', state.status);+ const qs = q.toString();+ history.replaceState(null, '', `${location.pathname}${qs ? `?${qs}` : ''}`);+}++function syncFilter() {+ filterEl.querySelectorAll('button').forEach((b) => b.setAttribute('aria-pressed', String(b.dataset.status === state.status)));+ if (!state.experiments) return;+ for (const s of STATUSES) {+ const n = s === 'all' ? state.experiments.length : state.experiments.filter((e) => e.status === s).length;+ const el = filterEl.querySelector(`[data-count="${s}"]`);+ if (el) el.textContent = String(n);+ }+}++function card(e) {+ const links = [`<a class="is-primary" href="${escapeHtml(e.url)}">Read the write-up</a>`];+ if (e.live_url) links.push(`<a href="${escapeHtml(e.live_url)}">Live view</a>`);+ for (const r of e.related || []) links.push(`<a href="${escapeHtml(r.href)}">${escapeHtml(r.label)}</a>`);+ return `<article class="ex-card">+ <div class="ex-meta">+ <span class="ex-status" data-status="${escapeHtml(e.status)}">${escapeHtml(STATUS_LABEL[e.status] || e.status)}</span>+ <span>Published <time datetime="${escapeHtml(e.published)}">${escapeHtml(fmtDate(e.published))}</time></span>+ <span>Spend ${escapeHtml(fmtSpend(e.spend_usd))}</span>+ </div>+ <h2><a href="${escapeHtml(e.url)}">${escapeHtml(e.title)}</a></h2>+ <p class="ex-q"><strong>Question:</strong> ${escapeHtml(e.question)}</p>+ ${e.headline ? `<p class="ex-headline">${escapeHtml(e.headline)}</p>` : ''}+ <div class="ex-links">${links.join('')}</div>+ </article>`;+}++function render() {+ syncFilter();+ listEl.removeAttribute('aria-busy');+ const all = state.experiments || [];+ if (!all.length) {+ listEl.innerHTML = `<div class="ex-empty"><strong>No experiments published yet.</strong>The first write-up lands here as soon as it is measured. <a href="/docs/experiments">See how an experiment is run and published</a>.</div>`;+ return;+ }+ const shown = state.status === 'all' ? all : all.filter((e) => e.status === state.status);+ if (!shown.length) {+ listEl.innerHTML = `<div class="ex-empty"><strong>No ${escapeHtml(STATUS_LABEL[state.status].toLowerCase())} experiments right now.</strong><button type="button" data-show-all>Show all ${all.length} experiments</button></div>`;+ listEl.querySelector('[data-show-all]').addEventListener('click', () => setStatus('all'));+ return;+ }+ listEl.innerHTML = shown.map(card).join('');+}++function renderError(detail) {+ listEl.removeAttribute('aria-busy');+ listEl.innerHTML = `<div class="ex-error" role="alert"><span><strong>Could not load the experiment log.</strong> ${escapeHtml(detail)}</span><button type="button" data-retry>Retry</button></div>`;+ listEl.querySelector('[data-retry]').addEventListener('click', load);+}++function setStatus(s) {+ state.status = s;+ writeUrl();+ render();+}++async function load() {+ listEl.setAttribute('aria-busy', 'true');+ listEl.innerHTML = '<div class="ex-skel skeleton" aria-hidden="true"></div><div class="ex-skel skeleton" aria-hidden="true"></div>';+ let body;+ try {+ const res = await fetch(INDEX_URL, { headers: { accept: 'application/json' } });+ if (!res.ok) throw new Error(`http_${res.status}`);+ body = await res.json();+ if (!Array.isArray(body?.experiments)) throw new Error('bad_body');+ } catch (err) {+ const status = /^http_(\d+)$/.exec(err?.message || '')?.[1];+ renderError(status ? `The server answered HTTP ${status}.` : err?.message === 'bad_body' ? 'The log could not be read.' : 'Network error: check your connection, then retry.');+ return;+ }+ state.experiments = body.experiments;+ render();+}++filterEl.addEventListener('click', (e) => {+ const b = e.target.closest('button[data-status]');+ if (!b || b.dataset.status === state.status) return;+ setStatus(b.dataset.status);+});+window.addEventListener('popstate', () => {+ readUrl();+ if (state.experiments) render();+ else syncFilter();+});++readUrl();+syncFilter();+load();diff --git a/tests/experiments-index.test.js b/tests/experiments-index.test.jsnew file mode 100644index 000000000..02d666779--- /dev/null+++ b/tests/experiments-index.test.js@@ -0,0 +1,92 @@+// The public experiment log (/experiments): the validator that gates+// `npm run build:pages`, and the index it publishes. The real data file is+// validated too, so a broken entry fails `npm test` as well as the build.++import { describe, it, expect } from 'vitest';+import { resolve } from 'node:path';+import {+ validateExperiments,+ buildExperimentsJson,+ loadExperiments,+ WRITEUP_SECTIONS,+} from '../scripts/lib/experiments-index.mjs';++const GOOD_DOC = `# Title\n\n${WRITEUP_SECTIONS.map((s) => `## ${s}\n\nBody.\n`).join('\n')}\n## Appendix: the queries\n`;++function entry(overrides = {}) {+ return {+ slug: 'agent-launch-funnel',+ title: 'Funnel',+ question: 'Who launches?',+ published: '2026-09-30',+ spend_usd: 0,+ status: 'concluded',+ doc: 'experiments-agent-launch-funnel',+ ...overrides,+ };+}++const docs = (map) => ({ readDoc: (doc) => (doc in map ? map[doc] : null) });++describe('validateExperiments', () => {+ it('accepts a complete entry whose write-up has the six sections in order', () => {+ const out = validateExperiments({ experiments: [entry()] }, docs({ 'experiments-agent-launch-funnel': GOOD_DOC }));+ expect(out).toHaveLength(1);+ });++ it.each(['slug', 'title', 'question', 'published', 'spend_usd', 'status', 'doc'])('fails on a missing %s', (field) => {+ const e = entry();+ delete e[field];+ expect(() => validateExperiments({ experiments: [e] }, docs({ 'experiments-agent-launch-funnel': GOOD_DOC }))).toThrow(+ field === 'doc' || field === 'slug' ? /missing required field|doc must be/ : new RegExp(`missing required field "${field}"`),+ );+ });++ it('fails on an unknown status, a negative spend and a bad date', () => {+ const io = docs({ 'experiments-agent-launch-funnel': GOOD_DOC });+ expect(() => validateExperiments({ experiments: [entry({ status: 'paused' })] }, io)).toThrow(/status must be one of/);+ expect(() => validateExperiments({ experiments: [entry({ spend_usd: -1 })] }, io)).toThrow(/spend_usd/);+ expect(() => validateExperiments({ experiments: [entry({ spend_usd: '5' })] }, io)).toThrow(/spend_usd/);+ expect(() => validateExperiments({ experiments: [entry({ published: '2026-13-45' })] }, io)).toThrow(/published/);+ });++ it('fails when the write-up does not exist', () => {+ expect(() => validateExperiments({ experiments: [entry()] }, docs({}))).toThrow(/does not exist/);+ });++ it('fails when a template section is missing or out of order', () => {+ const missing = GOOD_DOC.replace('## What we got wrong', '## Lessons');+ expect(() => validateExperiments({ experiments: [entry()] }, docs({ 'experiments-agent-launch-funnel': missing }))).toThrow(+ /"## What we got wrong"/,+ );+ const swapped = GOOD_DOC.replace('## Spend', '## TMP').replace('## Result', '## Spend').replace('## TMP', '## Result');+ expect(() => validateExperiments({ experiments: [entry()] }, docs({ 'experiments-agent-launch-funnel': swapped }))).toThrow(+ /out of order/,+ );+ });++ it('fails on a doc slug that does not follow experiments-<slug>, a duplicate slug and a bad live_url', () => {+ const io = docs({ 'experiments-agent-launch-funnel': GOOD_DOC, 'trading-experiment': GOOD_DOC });+ expect(() => validateExperiments({ experiments: [entry({ doc: 'trading-experiment' })] }, io)).toThrow(/doc must be/);+ expect(() => validateExperiments({ experiments: [entry(), entry()] }, io)).toThrow(/duplicate slug/);+ expect(() => validateExperiments({ experiments: [entry({ live_url: 'https://x.test' })] }, io)).toThrow(/live_url/);+ });+});++describe('buildExperimentsJson', () => {+ it('publishes newest first with the write-up URL', () => {+ const out = JSON.parse(+ buildExperimentsJson([entry({ slug: 'a', published: '2026-09-01' }), entry({ slug: 'b', published: '2026-09-30' })]),+ );+ expect(out.count).toBe(2);+ expect(out.experiments.map((e) => e.slug)).toEqual(['b', 'a']);+ expect(out.experiments[0].url).toBe('/docs/experiments-agent-launch-funnel');+ });+});++describe('the committed experiment log', () => {+ it('validates against the real write-ups', () => {+ const { entries } = loadExperiments(resolve(import.meta.dirname, '..'));+ expect(entries.length).toBeGreaterThan(0);+ });+});diff --git a/vercel.json b/vercel.jsonindex 644fcee8e..f7af5dc7b 100644--- a/vercel.json+++ b/vercel.json@@ -446,6 +446,10 @@"src": "/analytics/?","dest": "/analytics.html"},+ {+ "src": "/experiments/?",+ "dest": "/experiments.html"+ },{"src": "/economy","dest": "/economy.html"diff --git a/vite.config.js b/vite.config.jsindex eefb5c78c..81c1096b2 100644--- a/vite.config.js+++ b/vite.config.js@@ -869,6 +869,7 @@ const appConfig = {'agent-economy': resolve(__dirname, 'pages/agent-economy.html'),'agent-economy-volume': resolve(__dirname, 'pages/agent-economy-volume.html'),analytics: resolve(__dirname, 'pages/analytics.html'),+ experiments: resolve(__dirname, 'pages/experiments.html'),economy: resolve(__dirname, 'pages/economy.html'),'overlay-control': resolve(__dirname, 'pages/overlay-control.html'),'mocap-studio': resolve(__dirname, 'pages/mocap-studio.html'),