Promote dev to main: Serialized onboarding, bench scoring fixes, 001 node reference

.env.example

@@ -9,3 +9,9 @@
 # Optional. Live-page WebSocket relay. Public (NEXT_PUBLIC_) so it's baked
 # into the browser bundle. Defaults to the production relay if unset.
 # NEXT_PUBLIC_RELAY_WS_URL=wss://ocb-stream-relay-production.up.railway.app/ws
+
+# Optional. Minimum seconds between two full ISR tag purges triggered by
+# the materialize worker's /api/internal/revalidate-aggregate hook. The
+# worker calls the hook after every 60 s publish; without this throttle
+# every ISR page was invalidated each minute. Default 600. "0" disables.
+# REVALIDATE_MIN_INTERVAL_SEC=600

.github/workflows/ci.yml

@@ -26,3 +26,40 @@ jobs:
       - run: pnpm lint
       - run: pnpm test
       - run: pnpm build
+
+  # crm/ is its own package (Railway); the root typecheck excludes it, so it
+  # gets its own check. Runs only when the app changes.
+  crm:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: dorny/paths-filter@v3
+        id: changed
+        with:
+          filters: |
+            crm:
+              - 'crm/**'
+      - uses: pnpm/action-setup@v4
+        if: steps.changed.outputs.crm == 'true'
+        with:
+          version: 10
+      - uses: actions/setup-node@v4
+        if: steps.changed.outputs.crm == 'true'
+        with:
+          node-version: 22
+      - uses: oven-sh/setup-bun@v2
+        if: steps.changed.outputs.crm == 'true'
+        with:
+          bun-version: latest
+      - run: pnpm install --frozen-lockfile --ignore-workspace
+        if: steps.changed.outputs.crm == 'true'
+        working-directory: crm
+      - run: pnpm typecheck
+        if: steps.changed.outputs.crm == 'true'
+        working-directory: crm
+      - run: pnpm test
+        if: steps.changed.outputs.crm == 'true'
+        working-directory: crm
+      - run: pnpm build
+        if: steps.changed.outputs.crm == 'true'
+        working-directory: crm

.github/workflows/prod-deploy.yml

@@ -30,6 +30,12 @@ jobs:
       CRON_SECRET: ${{ secrets.CRON_SECRET }}
     steps:
       - uses: actions/checkout@v4
+        with:
+          # Full history: scripts/emit-page-mtimes.ts (prebuild) reads each
+          # content file's last commit for the sitemap <lastmod>. A depth-1
+          # checkout dates only the files touched by HEAD and every other
+          # URL falls back to the build time (929 of 929 on 2026-09-19).
+          fetch-depth: 0
 
       - uses: actions/setup-node@v4
         with:
@@ -48,6 +54,14 @@ jobs:
         run: vercel pull --yes --environment=production --token=${{ secrets.VERCEL_TOKEN }}
 
       - name: Build
+        # NEXT_PUBLIC_POSTHOG_KEY comes from a repository variable: `vercel pull`
+        # writes the literal "[SENSITIVE]" for env vars flagged sensitive on
+        # Vercel, and that string was compiled into the bundle as the PostHog
+        # token (no event reached PostHog from 2026-08-24 to 2026-09-20).
+        # process.env wins over the pulled .env file. The token is public by
+        # nature (it ships to every browser), so a variable, not a secret.
+        env:
+          NEXT_PUBLIC_POSTHOG_KEY: ${{ vars.NEXT_PUBLIC_POSTHOG_KEY }}
         run: vercel build --prod --token=${{ secrets.VERCEL_TOKEN }}
 
       - name: Deploy to Production

.github/workflows/staging-deploy.yml

@@ -23,6 +23,12 @@ jobs:
       VERCEL_PROJECT_ID: ${{ secrets.VERCEL_PROJECT_ID }}
     steps:
       - uses: actions/checkout@v4
+        with:
+          # Full history: scripts/emit-page-mtimes.ts (prebuild) reads each
+          # content file's last commit for the sitemap <lastmod>. A depth-1
+          # checkout dates only the files touched by HEAD and every other
+          # URL falls back to the build time (929 of 929 on 2026-09-19).
+          fetch-depth: 0
 
       - uses: actions/setup-node@v4
         with:
@@ -41,6 +47,10 @@ jobs:
         run: vercel pull --yes --environment=preview --token=${{ secrets.VERCEL_TOKEN }}
 
       - name: Build
+        # Same repository variable as prod-deploy.yml (see the note there).
+        # Staging events are excluded by the CRM's host filter.
+        env:
+          NEXT_PUBLIC_POSTHOG_KEY: ${{ vars.NEXT_PUBLIC_POSTHOG_KEY }}
         run: vercel build --token=${{ secrets.VERCEL_TOKEN }}
 
       - name: Deploy Preview

README.md

@@ -62,12 +62,12 @@ Source: OpenChainBench (https://openchainbench.com/benchmarks/aggregator-head-la
 | [`/`](https://openchainbench.com/) | Hero + Latest benchmarks table + Live dashboard |
 | [`/benchmarks`](https://openchainbench.com/benchmarks) · `/benchmarks/[slug]` | Catalog grid + per-bench leaderboard with chart, ledger, chain/region filters, share cards, FAQ |
 | [`/benchmarks/[slug]/[chain]`](https://openchainbench.com/benchmarks/aggregator-head-lag/solana) | Chain-scoped bench view with per-chain explainer |
-| [`/products/[slug]`](https://openchainbench.com/products/mobula) | Per-provider aggregated profile: every bench they appear in, top-1 counts, related products |
+| [`/products/[slug]`](https://openchainbench.com/products/mobula) | The one page per product: every bench it appears in, top-1 counts, related products, and a view per category it belongs to (Trading app, Perpetuals, Prediction markets, Data API, Bridge, RPC, Hyperliquid) selectable by pill or URL hash |
 | [`/chains/[slug]`](https://openchainbench.com/chains/ethereum) | Per-chain hub: live native price, TVL history, every bench that touches this chain grouped by category |
 | [`/compare/[a]-vs-[b]`](https://openchainbench.com/compare/codex-vs-mobula) | Head-to-head comparison across shared benches, canonical alphabetical order enforced at the edge |
 | [`/alternatives/[slug]`](https://openchainbench.com/alternatives/alchemy) | "Alternatives to X" landing pages with live leaderboard from the referenced bench |
 | [`/answers/[slug]`](https://openchainbench.com/answers/which-crypto-data-api-covers-the-most-blockchains) | Q&A pages backed by a live bench, formatted for AI answer engines |
-| [`/hyperliquid`](https://openchainbench.com/hyperliquid) · `/hyperliquid/[slug]` | HyperLiquid frontends cohort + per-builder dashboard (revenue, volume, first-active date) |
+| [`/hyperliquid`](https://openchainbench.com/hyperliquid) | HyperLiquid frontends cohort; each builder's dashboard (revenue, volume, first-active date) is the Hyperliquid view on its `/products/[slug]` page |
 | [`/perps`](https://openchainbench.com/perps) · [`/prediction-markets`](https://openchainbench.com/prediction-markets) · [`/rpc`](https://openchainbench.com/rpc) | Vertical hubs aggregating multiple benches per topic |
 | [`/mcp`](https://openchainbench.com/mcp) | MCP server docs + install instructions |
 | [`/methodology`](https://openchainbench.com/methodology) · [`/contribute`](https://openchainbench.com/contribute) · [`/team`](https://openchainbench.com/team) · [`/about`](https://openchainbench.com/about) · [`/press`](https://openchainbench.com/press) · [`/badges`](https://openchainbench.com/badges) | Static |

answers/cheapest-bridge-usdc-to-base.yml

@@ -1,41 +1,42 @@
 slug: cheapest-bridge-usdc-to-base
 question: "Which bridge is cheapest to move USDC to Base?"
 short_answer: |
-  OpenChainBench measures the all in cost of a $300 USDC bridge (fee plus slippage plus destination gas) across eight routers: Mobula, Relay, LI.FI, deBridge, Across, Near Intents, Squid and Socket. Since the bench aggregates across the Solana, Base and Arbitrum corridors, the USDC-to-Base leader on a specific corridor is on the per corridor breakdown at openchainbench.com/benchmarks/bridge-fee.
+  {{best_name:chain:Base}} is the cheapest measured bridge for $300 USDC from Solana to Base at {{best_p50:chain:Base}} all in (fee plus slippage plus destination gas, p50 over 24h); {{worst_name:chain:Base}} is the most expensive on that corridor at {{worst_p50:chain:Base}}. Across every corridor the leader is {{best_name}} at {{best_p50}}, measured live by OpenChainBench across {{count}} routers.
 
 benchmark: bridge-fee
 
 intro: |
-  Bridge comparison articles publish rack rate fee schedules and stop there. The rack rate is one number in a bill that also includes destination gas, LP fees passed through as slippage, and the effective execution price the router actually delivers. On a $300 USDC bridge to Base, two providers quoting the same 5 bps headline can deliver very different real fees once the router walks the pool. This page answers the specific question retail and small-treasury movers actually search for: at $300 USDC notional, which bridge to Base is cheapest today. OpenChainBench queries the same USDC to Base intent against every major router every 5 minutes, computes the all-in effective fee including destination gas, and publishes the ranked leaderboard live.
+  Bridge comparison articles publish rack rate fee schedules and stop there. The rack rate is one number in a bill that also includes destination gas, LP fees passed through as slippage, and the effective execution price the router actually delivers. On a $300 USDC bridge to Base, two providers quoting the same 5 bps headline can deliver very different real fees once the router walks the pool. This page answers the specific question retail and small treasury movers actually search for: at $300 USDC notional, which bridge into Base is cheapest today. OpenChainBench sends the same Solana to Base USDC quote request to every measured router every 5 minutes, computes the all in cost including destination gas, and publishes the ranked leaderboard live; the Sol to Base tab of the bench page is the corridor this answer reads.
 
 methodology: |
-  Every 5 minutes the harness sends the same intent (source: Ethereum mainnet USDC, destination: Base USDC, amount: $300 notional, receiver: probe wallet) to each router's quote API in parallel. Effective fee is computed as `(source_input_usd - destination_output_usd - destination_gas_usd) / source_input_usd` in basis points, so slippage, LP fees and gas are all inside the same number. Providers that fail to return a quote (5xx, rate limit, timeout) leave the previous gauge in place and increment a fetch_errors_total counter. The leaderboard ranks by 24h average of bridge_effective_fee_bps for the USDC to Base pair. The $300 size is the small ticket regime; sibling benches probe higher notional to expose where routers slip on size.
+  Every 5 minutes the bridge-monitor harness requests a quote for the same intent (source: USDC on Solana, destination: USDC on Base, amount: $300) from each router's quote API. The recorded metric is `bridge_cost_percent`, the share of the $300 ticket that does not arrive on Base once the explicit fee, the slippage and the destination gas returned by the quote are summed. Quotes that error (unsupported route, quote_failed, timeout) are excluded from the cost aggregate and counted toward the success rate, so a broken router cannot post a zero. The leaderboard ranks by the 24h p50 of `bridge_cost_percent{bridge="X", amount_usd="300", to_chain="Base"}`; p90 and p99 capture the worst minutes. The bench page's Sol to Base tab is this corridor; the other tabs (Base to Arb, Arb to Sol, Arb to HyperCore) show the same routers on the other measured routes.
 
 limitations:
-  - "Quote fee only. This measures router-quoted effective fee, not settlement. Any router that quotes tighter than it settles would show a fast p50 with degraded real-world execution; the sibling bridge-quote-latency bench flags routers where quote-to-settle drift is measurable."
-  - "One size, one route. The $300 USDC to Base pair is the highest-search-volume small-ticket bridge query. At $50k notional the leaderboard re-orders because the spread and LP fee components dominate. Producers moving treasury size should read the per-size breakdown on the bench page."
-  - "Destination gas at quote time. Destination gas is included in the effective fee at the quoted receiver payout, not at a delayed execution instant. A destination gas spike between quote and settle would degrade the actual fee below what this leaderboard reports."
-  - "No canonical CCTP fast path. The USDC canonical bridge via Circle CCTP has different guarantees and speeds than the market-maker routers ranked here. The leaderboard measures the market-maker path (Mobula, Relay, LI.FI, deBridge, Across, Near Intents) because it is the head-to-head every comparison article ranks."
+  - "Quote cost, not settled cost. This measures what the router quotes for the transfer, not what lands after settlement. A router that quotes tighter than it settles would look cheaper here than in a wallet; a separate execution bench on the same harness reads the realized amount on chain for the routers that support programmatic execution."
+  - "One size, one route. The $300 USDC Solana to Base corridor is the highest search volume small ticket bridge query. At $10000 the ranking re-orders because the fixed components (destination gas, base protocol fee) dilute into the percentage; the harness quotes $5, $50 and $300 only, so larger tickets are not measured."
+  - "Destination gas at quote time. Destination gas is included in the all in cost at the quoted payout, not at a delayed execution instant. A Base gas spike between quote and settlement would move the real cost away from what this leaderboard reports."
+  - "No canonical CCTP path. Circle's burn and mint bridge has different guarantees and speeds than the market maker routers ranked here. The leaderboard measures the router path because it is the head to head every comparison article ranks."
+  - "Single point of measurement in EU-West (Paris). A router whose solvers sit in us-east can quote differently from there; the us-east and Singapore probes are paused."
 
 faq:
   - q: "Which bridge is cheapest for a $300 USDC transfer to Base?"
-    a: "The bench-fee leaderboard on the Base corridor is at openchainbench.com/benchmarks/bridge-fee. Eight routers compete on the same $300 USDC probe (Mobula, Relay, LI.FI, deBridge, Across, Near Intents, Squid and Socket) and the ranking re-sorts every 5 minutes against fresh Prometheus samples."
+    a: "{{best_name:chain:Base}} at {{best_p50:chain:Base}} all in on the Solana to Base corridor (p50 over 24h). {{count}} routers compete on the same $300 USDC quote and the ranking re-sorts every 5 minutes against fresh Prometheus samples; the most expensive on this corridor is {{worst_name:chain:Base}} at {{worst_p50:chain:Base}}."
   - q: "Why $300, and does the winner change at $10k or $100k?"
-    a: "$300 is the small-ticket regime that dominates retail search. At $10k the leaderboard shifts because aggregator markups compress against intent-based routers (Across, Near Intents, Mobula) whose quotes are less size-dependent. At $100k the spread widens further and the choice becomes size-specific; the per-size tab on the bench page shows the leader at each notional."
+    a: "$300 is the small ticket regime that dominates retail search. At $10k the ranking shifts because fixed components (destination gas, protocol base fee) dilute into the percentage and aggregator markups compress against intent based routers whose quotes are less size dependent. The harness quotes $5, $50 and $300, so this page cannot name the leader at $10k; treasury size transfers should be quoted directly."
   - q: "Is Circle CCTP not on this ranking?"
-    a: "CCTP is Circle's canonical burn-and-mint path for USDC. It has different guarantees, speeds and a fixed protocol path, so it is not directly comparable to market-maker routers that price a swap through a pool. The bench measures the market-maker rankings because that is the choice retail actually makes; canonical CCTP is a separate decision users make once they know their maximum acceptable latency."
-  - q: "How is Across usually so cheap on this route?"
-    a: "Across runs an intent-based settlement model where relayers compete to fill from destination inventory. That model compresses the spread component to near zero on high-liquidity pairs like USDC Ethereum to Base. On less-liquid destinations Across loses its edge; that per-destination effect is why the leaderboard re-ranks by pair rather than emitting a single global cheapest bridge."
+    a: "CCTP is Circle's canonical burn and mint path for USDC. It has different guarantees, speeds and a fixed protocol path, so it is not directly comparable to market maker routers that price a transfer through a pool or a solver auction. The bench measures the router rankings because that is the choice retail actually makes; canonical CCTP is a separate decision users make once they know their maximum acceptable latency."
+  - q: "Why does the Base corridor leader differ from the overall leader?"
+    a: "The overall headline pools the Solana, Base, Arbitrum and HyperCore corridors. A solver network with deep inventory on Base can lead Sol to Base and trail on Arb to HyperCore, where inventory is thinner. Today the overall leader is {{best_name}} at {{best_p50}} and the Sol to Base leader is {{best_name:chain:Base}} at {{best_p50:chain:Base}}; the two names coincide only when one router leads on every route."
   - q: "Which USDC routers are on this leaderboard?"
-    a: "Eight routers: Mobula, Relay, LI.FI, deBridge, Across, Near Intents, Squid and Socket. Stargate is not measured because its $300 USDC quote API did not meet the harness's coverage or reliability threshold."
+    a: "{{count}} routers with live data today. The measured set spans direct protocols, aggregators, intent layers and relays; a router drops out of the ranking when its quotes fail for the whole 24h window and returns when they resume. Stargate is not measured because its $300 USDC quote API did not meet the harness's coverage or reliability threshold."
   - q: "How often is the leaderboard refreshed?"
-    a: "Every 5 minutes. The harness probes every router's quote API at that cadence, publishes bridge_effective_fee_bps to Prometheus, and the page reads a 24h rolling average. Because the recording rules re-aggregate every 30 seconds the on-screen ranking re-orders within a minute of any router changing quote behavior."
+    a: "Every 5 minutes. The harness requests a quote from every router at that cadence, publishes `bridge_cost_percent` to Prometheus, and the page reads a rolling 24h p50, p90 and p99, so a single off market quote cannot move the headline."
 
 related:
   - which-bridge-has-the-fastest-quote-api
   - which-bridge-has-the-cheapest-usdc-fee
   - which-blockchain-has-cheapest-transaction-fees
 
-seo_title: "Cheapest bridge for USDC to Base 2026 live"
-seo_description: "OpenChainBench ranks six USDC routers (Mobula, Relay, LI.FI, deBridge, Across, Near Intents) on all in cost at $300 notional across Solana, Base and Arbitrum. Live per corridor breakdown on the bench page."
+seo_title: "Cheapest bridge for USDC to Base: {{best_name:chain:Base}} at {{best_p50:chain:Base}}"
+seo_description: "{{best_name:chain:Base}} is the cheapest bridge for $300 USDC from Solana to Base at {{best_p50:chain:Base}} all in (fees, slippage, gas), measured live every 5 minutes across {{count}} routers."
 status: live

answers/how-much-does-it-cost-to-bridge-usdc-to-arbitrum.yml

@@ -33,6 +33,6 @@ related:
   - cheapest-bridge-usdc-to-base
   - which-bridge-has-the-fastest-quote-api
 
-seo_title: "How much does it cost to bridge USDC to Arbitrum? Live all in fees"
+seo_title: "How much does it cost to bridge USDC to Arbitrum? Live fees"
 seo_description: "Bridging $300 of USDC costs from {{best_p50}} to {{worst_p50}} all in (p50, 24h) depending on provider. deBridge, LI.FI, Mobula, Relay, Across and Near Intents measured live by OpenChainBench."
 status: live

answers/is-polymarket-legal-in-the-us.yml

@@ -83,5 +83,5 @@ related:
   - best-prediction-market-for-politics
   - which-prediction-market-data-api-is-the-freshest
 
-seo_title: "Is Polymarket legal in the US? CFTC settlement, QCX and state bans explained"
+seo_title: "Is Polymarket legal in the US? CFTC settlement and state bans"
 seo_description: "Polymarket.com geoblocked US users after a 2022 CFTC settlement. Polymarket US (QCX) launched December 2025 as the CFTC-regulated US option. Kalshi is also CFTC-regulated. Minnesota banned prediction markets Aug 2026."