Prepare Pareto FalconX solver onboarding

README.md

@@ -82,6 +82,12 @@ use one candidate route. Other scopes keep the existing multi-route behavior.
 `minAmountsIn` adds an optional per-input-token floor on request size (base units, decimal strings):
 a request below its token's minimum is not quoted (HTTP 204), while an amount equal to the minimum
 still quotes; unlisted tokens have no floor.
+Pareto's mainnet `AA_FalconXUSDC` tranche (`0xC26A…f99C`) uses this existing generic path and needs no
+token-specific solver code. On an instance already running `tokensToQuote: permissioned` — which is
+what admits by `permissionedTokens` at all, since the default `all` scope admits every token
+regardless — onboarding it is the same two config edits as any other permissioned token: add it to
+`permissionedTokens`, and give it a base-unit floor in `minAmountsIn`. Quoting it also requires a
+LiquidLane adapter holding the token in `adapters`.
 When an exact-input request exceeds the advertised adapter capacity, the default strategy caps the
 quoted output at the available `maxAssets` instead of declining in every token scope; the excess input
 is reflected as worse execution price and price impact. Awarded orders are planned again from current

config/rfq.example.yaml

@@ -67,12 +67,14 @@ solvers:
       # tokensToQuote: all
       # permissionedTokens:            # with tokensToQuote: permissioned, inputs must use one candidate route
       #   - "0x..."
+      #   - "0xC26A6Fa2C37b38E549a4a1807543801Db684f99C"  # Pareto AA_FalconXUSDC (mainnet)
 
       # minAmountsIn sets a per-input-token floor on request size, in that token's BASE UNITS (decimal
       # string). A request below its token's minimum gets no quote (HTTP 204); an amount equal to the
       # minimum still quotes. Tokens not listed have no minimum.
       # minAmountsIn:
       #   "0x238a700eD6165261Cf8b2e544ba797BC11e466Ba": "1000000000000000000"  # 1 token (18 decimals)
+      #   "0xC26A6Fa2C37b38E549a4a1807543801Db684f99C": "1000000000000000000"  # 1 AA_FalconXUSDC
 
       # LiquidLane adapter instances this filler serves — the concrete per-token adapters (not the
       # factory). Each adapter's vault + asset are resolved on-chain at startup. In `external` mode

docs/RFQ-PLAN.md

@@ -289,6 +289,26 @@ refresh uses (`paused`, `getMaxAssets`, `getMaxRate`) — each adapter's `vault`
 
 ## 5. Open items / prerequisites
 
+- **Pareto AA_FalconXUSDC readiness** — the RFQ implementation is already token-agnostic and supports
+  the 18-decimal mainnet tranche `0xC26A6Fa2C37b38E549a4a1807543801Db684f99C` through the existing
+  permissioned, single-route path. Deployed and verified on mainnet as of 2026-08-06: the bounded
+  Symbiotic `ParetoOracle` (`0xba833D6288aC591BFffbeD16909B8B824e7fA9F2`, block 25697403, already
+  wired into the rfq-backend and rfq-frontend token entries) and the Pareto account
+  factory/implementation (`0xa47cE86c304e251198F1a318c569e5217b694e1F` /
+  `0xE54219880a1296D028CcD7B3d5a34c65fb0B7CAB`). Still outstanding, because each is an instance
+  someone must create rather than an address to configure: the Keyring-onboarded account proxy, the
+  LiquidLane adapter, the RFQ Executor's authorized caller, and backend token enablement. At
+  activation, on an instance already scoped `tokensToQuote: permissioned` (the default `all` scope
+  admits every token, so `permissionedTokens` governs nothing there), configure the token in
+  `permissionedTokens`, set its 1-token floor (`1000000000000000000`) in `minAmountsIn`, and point
+  the instance at the LiquidLane adapter holding it via `adapters`. No dedicated per-token process
+  exists: separate solver instances differ only by executor and settlement adapter, not by asset.
+  The checked-in RFQ example carries the exact token/floor snippet; no protocol branch belongs in
+  solver code.
+
+  Deployment status for this item lives here rather than in the README, which `AGENTS.md` reserves
+  for the external operator-facing runtime and configuration surface.
+
 - **Authorized caller of the `Executor`** — the bot EOA must be added to the Executor's `callers`
   allowlist (owner-only `setCallers`) before fills land (onboarding
   prereq, analogous to 3F's offer-signer). Document; do not grant from the bot.