chore: scrub internal session-log narrative from code comments

Wide sweep across the codebase to remove leftover artifacts of internal
development sessions, internal entity naming, and audit-code references
that point at non-public docs. The technical reasoning for each piece
of code stays; the "Caught 2026-05-XX while debugging XYZ at preprod"
narrative goes.

Categories scrubbed:
- Dated session-log comments ("Caught/Surfaced/Discovered 2026-05-XX")
  → rewritten as neutral technical reasoning.
- Internal audit codes (AUDIT-H2, AUDIT-C2, AUDIT-M2, AUDIT-H5, etc.)
  referencing a non-public audit doc → labels stripped, fix reasoning
  kept.
- Internal-entity names in code comments (Sulkta-specific, Sulkta runs
  X, Terrapin/TRP as gov-token names) → generic phrasing.
- Test fixture helper `sulkta_cfg` → `test_dao_cfg`; test DAO name
  string `"sulkta"` → `"test-dao"`. On-chain addresses in test fixtures
  kept (they're real-world wire-byte test data on public chain).
- Cross-references to memory files / non-public audit docs
  (`internal notes`, `aiken-escrow/README.md`)
  → reasoning inlined or removed.
- Test names renamed: `decodes_sulkta_live_governor_datum` →
  `decodes_live_governor_datum`, `decodes_sulkta_live_proposal_zero` →
  `decodes_live_finished_proposal`, etc.

Kept (legitimate):
- Cross-references to in-repo audit docs (aiken-escrow/README.md, aiken-escrow/README.md) — they ARE the
  public artifacts being referenced.
- HIGH-1/HIGH-2/MED-2/LOW labels on escrow fixes — these correspond to
  findings in the in-repo audit doc.
- TODO markers — legitimate work-still-to-do.

crates/aldabra-core/src/governance.rs

@@ -38,9 +38,9 @@ use crate::{Network, PaymentKey, ProtocolParams, StakeKey, WalletError};
 /// Conway DRep registration deposit. Mainnet protocol parameter
 /// `drep_deposit` is currently 500 ADA. **Use `params.drep_deposit_lovelace`
 /// instead of this constant** — it's kept here for backward-compat callers
-/// only. AUDIT-2026-05-06 M-2: hardcoding the deposit means a protocol
-/// change (or an old DRep registered at a different deposit) will silently
-/// fail ledger validation. Always pull from current chain params.
+/// only. Hardcoding the deposit means a protocol change (or an old DRep
+/// registered at a different deposit) will silently fail ledger validation.
+/// Always pull from current chain params.
 pub const DREP_REGISTRATION_DEPOSIT_LOVELACE: u64 = 500_000_000;
 
 /// Two witnesses (payment + stake) — same overhead as

crates/aldabra-core/src/plutus.rs

@@ -148,12 +148,11 @@ pub fn build_signed_plutus_spend(
 
     // AUDIT4-2 fix: pick the SMALLEST ADA-only UTXO that still
     // qualifies for collateral (≥ 5 ADA), so the LARGEST stays
-    // available for funding the spend. Previously we did the
-    // inverse — collateral got the biggest utxo, funding got
-    // whatever scrap was next, and a typical wallet (one big
-    // change utxo + a tiny self-send leftover) couldn't cover
-    // payout + fee + min_utxo even with billions of lovelace
-    // sitting in the change. Surfaced 2026-05-04 audit-4 phase F2.
+    // available for funding the spend. The inverse approach (give
+    // collateral the biggest utxo) breaks the common case where a
+    // wallet has one large change utxo + a small self-send leftover,
+    // since funding ends up with the scrap and can't cover payout +
+    // fee + min_utxo.
     //
     // Collateral is NEVER consumed on the happy path — it's only
     // seized if the script fails — so its size beyond the 5-ADA
@@ -544,10 +543,9 @@ mod tests {
 
     /// AUDIT4-2 regression. A wallet with one tiny qualifying UTXO
     /// alongside one huge UTXO must pick the tiny one for collateral
-    /// and the huge one for funding (not the inverse). Pre-fix, the
-    /// huge UTXO became collateral and funding fell back to the
-    /// tiny 5-ADA scrap, too small to cover payout, script-exec
-    /// fee, and change min_utxo. Surfaced 2026-05-04 audit-4 phase F2.
+    /// and the huge one for funding (not the inverse). The inverse
+    /// fails in the common wallet shape where funding then can't
+    /// cover payout + script-exec fee + change min_utxo.
     #[test]
     fn picks_smallest_qualifying_collateral_largest_funding() {
         let payment = payment_from_canonical();

crates/aldabra-core/src/plutus_cost_models.rs

@@ -1,9 +1,8 @@
-// Conway-era Plutus V3 cost model, 297 params. Snapshot from preprod
-// epoch 286 (2026-05) but **identical to mainnet epoch 629** —
-// confirmed 2026-05-04 by parallel Koios `epoch_params` fetch from
-// `api.koios.rest` and `preprod.koios.rest`. Cost models are
-// protocol-version parameters, not network parameters; they only
-// diverge if a network does an experimental hard fork off-cycle.
+// Conway-era Plutus V3 cost model, 297 params. Snapshot — verified
+// identical between preprod and mainnet via parallel Koios
+// `epoch_params` fetches. Cost models are protocol-version
+// parameters, not network parameters; they only diverge if a network
+// does an experimental hard fork off-cycle.
 //
 // Used by both preprod and mainnet Plutus paths today. Re-snapshot
 // from mainnet Koios after any major hard fork. If preprod and

crates/aldabra-core/src/plutus_mint.rs

@@ -19,7 +19,7 @@
 //! - **Mint**: caller-supplied `(asset_name_hex, quantity)` list under
 //!   the supplied policy (Plutus V1/V2/V3).
 //! - **Recipient output**: address + lovelace + minted assets +
-//!   any caller-supplied extra assets to forward (e.g. tTRP gov tokens
+//!   any caller-supplied extra assets to forward (e.g. gov tokens
 //!   on a stake bootstrap) + optional inline datum.
 //! - **Change output**: leftover ADA + leftover input assets (other
 //!   than what was forwarded to the recipient).
@@ -29,7 +29,7 @@
 //! Agora's deployment pattern is the same shape for every "first-time
 //! mint of a single ST token under a Plutus policy" tx:
 //! - Governor bootstrap: mint 1 GST → governor_addr + GovernorDatum
-//! - Stake bootstrap: mint 1 StakeST → stakes_addr + tTRP + StakeDatum
+//! - Stake bootstrap: mint 1 StakeST → stakes_addr + gov-token + StakeDatum
 //! - Proposal create: mint 1 ProposalST → proposal_addr + ProposalDatum
 //!
 //! All three share the structure; the only differences are the
@@ -57,7 +57,7 @@ pub struct PlutusMintAsset {
 }
 
 /// Optional non-mint asset to attach to the recipient output.
-/// Used for e.g. "send tTRP alongside the freshly-minted StakeST"
+/// Used for e.g. "send gov-tokens alongside the freshly-minted StakeST"
 /// on a stake bootstrap. Sourced from wallet input UTxOs.
 #[derive(Debug, Clone)]
 pub struct ExtraDestAsset {
@@ -90,7 +90,7 @@ pub struct PlutusMintArgs<'a> {
     pub dest_lovelace: u64,
     /// Non-mint assets to include on the recipient output. Sourced
     /// from wallet inputs. Empty for governor bootstrap; non-empty
-    /// for stake bootstrap (tTRP forwarded into the stake).
+    /// for stake bootstrap (gov-tokens forwarded into the stake).
     pub dest_extra_assets: &'a [ExtraDestAsset],
     /// Optional inline datum on the recipient output. Required for
     /// any send to a Plutus script address.
@@ -101,9 +101,10 @@ pub struct PlutusMintArgs<'a> {
     /// for any `pauthorizedBy` / `txSignedBy` check inside a script.
     /// MCP layer always passes this wallet's payment-key pkh; pass
     /// extra entries for cosigners. Empty slice = scripts that don't
-    /// check signatories (e.g. Agora's GST policy).
-    /// Caught 2026-05-07 on Agora's stake-policy bootstrap on preprod
-    /// — script erred because owner pkh was absent from signatories.
+    /// check signatories (e.g. Agora's GST policy). Omitting a pkh
+    /// the script checks for will cause the script to error on
+    /// validation even though the corresponding VKey witness is
+    /// present.
     pub additional_signers: &'a [Hash<28>],
 }
 
@@ -595,11 +596,9 @@ fn prepare_plutus_mint(
             }
 
             // Plutus V1/V2/V3 each need their cost-model wired via
-            // language_view so pallas computes script_data_hash on the tx
-            // body. Without it, chain rejects with PPViewHashesDontMatch.
-            // Caught 2026-05-07 attempting Agora's V2 GST-policy bootstrap
-            // mint on preprod — earlier code only set language_view for
-            // V3 and every V2 mint hit the chain rejection.
+            // language_view so pallas computes script_data_hash on the
+            // tx body. Without it, chain rejects with
+            // PPViewHashesDontMatch.
             match args.policy_version {
                 PlutusVersion::V2 => {
                     staging = staging.language_view(
@@ -738,7 +737,7 @@ mod tests {
     }
 
     /// Sample preprod governor address (the one Plutarch linker
-    /// produced for our preprod tTRP DAO). Used as the dest.
+    /// produced for our preprod gov-token DAO). Used as the dest.
     const SAMPLE_GOVERNOR_ADDR: &str =
         "addr_test1wqlzsnytzs4qv0trmhvw5cuyxnxk0qjq68crn85jj4lhv7qn4wym4";
 

crates/aldabra-core/src/tx.rs

@@ -383,19 +383,18 @@ fn output_with_assets(
             .add_asset(policy, name, *qty)
             .map_err(|e| WalletError::Derivation(format!("output add_asset: {e}")))?;
     }
-    // AUDIT4-3 fix: optional inline datum for locking funds at a script
-    // address. Without this, sending to a script address creates an
-    // un-spendable utxo (Babbage/Conway require script-locked outputs
-    // to carry a datum). Caller passes the PlutusData CBOR of whatever
-    // shape the validator expects.
+    // Optional inline datum for locking funds at a script address.
+    // Without this, sending to a script address creates an un-spendable
+    // utxo (Babbage/Conway require script-locked outputs to carry a
+    // datum). Caller passes the PlutusData CBOR of whatever shape the
+    // validator expects.
     if let Some(datum) = inline_datum_cbor {
         out = out.set_inline_datum(datum.to_vec());
     }
-    // 2026-05-07: optional reference-script attached to the output.
-    // This is the on-chain equivalent of `cardano-cli ... --tx-out
-    // --tx-out-reference-script-file ...`. Once deployed, downstream
-    // txs can witness the script via `read_only_input` instead of
-    // inline-witnessing the full CBOR. Required for any DAO/dApp that
+    // Optional reference-script attached to the output — equivalent of
+    // `cardano-cli ... --tx-out-reference-script-file ...`. Once deployed,
+    // downstream txs can witness the script via `read_only_input` instead
+    // of inline-witnessing the full CBOR. Useful for any DAO/dApp that
     // wants to keep witness sizes manageable when validators are large.
     if let Some(rs) = reference_script {
         out = out.set_inline_script(rs.kind, rs.cbor.to_vec());
@@ -554,18 +553,13 @@ fn prepare_payment(
     // Mint paths typically have more lovelace headroom and won't
     // hit the pass1 floor; if a mint does run tight, the downstream
     // "insufficient funds for fee" error is informative.
-    // Was 500_000 — surfaced 2026-05-05 zeroing out the mainnet
-    // test wallet (1.8 ADA out of 2 ADA refused upstream).
     let fee_pass1: u64 = 200_000;
-    // AUDIT5-1: ada-only sends fold sub-min change into fee on the
-    // happy path (see line ~552 below — the `Some(c)` ADA-only arm),
-    // so the selector shouldn't insist on having `min_utxo_lovelace`
-    // worth of room for change. Pass 0 when there are no asset
-    // leftovers; assets-bearing sends still need real change to
-    // route the leftover policy IDs, so keep min_utxo_lovelace there.
-    // Surfaced 2026-05-05 trying to zero out the mainnet test wallet:
-    // 2 ADA balance, 1.8 ADA send refused as "need 3.3M, have 2M"
-    // even though the chain math was fine.
+    // ada-only sends fold sub-min change into fee on the happy path
+    // (the `Some(c)` ADA-only arm below), so the selector shouldn't
+    // insist on having `min_utxo_lovelace` worth of room for change.
+    // Pass 0 when there are no asset leftovers; assets-bearing sends
+    // still need real change to route the leftover policy IDs, so
+    // keep min_utxo_lovelace there.
     let min_change_required = if target_assets.is_empty() {
         0
     } else {
@@ -1286,11 +1280,10 @@ mod tests {
         assert_eq!(result.summary.change_assets[0].policy_id_hex, policy);
     }
 
-    /// AUDIT4-3 regression: a wallet_send with `to_inline_datum_cbor`
-    /// produces an output carrying that datum. Without this we'd lock
-    /// funds at script addresses with no datum, which Babbage/Conway
-    /// rejects on spend. Surfaced 2026-05-04 audit-4 phase F2 against
-    /// the always-succeeds Aiken validator.
+    /// Regression: a wallet_send with `to_inline_datum_cbor` produces
+    /// an output carrying that datum. Without this we'd lock funds at
+    /// script addresses with no datum, which Babbage/Conway rejects on
+    /// spend.
     #[test]
     fn lock_with_inline_datum_attaches_datum_to_output() {
         use pallas_primitives::Fragment;
@@ -1326,13 +1319,10 @@ mod tests {
         }
     }
 
-    /// AUDIT5-1 regression: ada-only sends should be allowed to drain
-    /// a wallet down to "all of input - fee" without the selector
-    /// reserving min_utxo for a change output that ends up folded
-    /// into the fee anyway. Pre-fix this returned "need 3300000
-    /// (target+fee+min_change), have 2000000" even though the chain
-    /// math is fine. Caught 2026-05-05 zeroing out the mainnet test
-    /// wallet during Phase 5 real-funds testing.
+    /// Regression: ada-only sends should be allowed to drain a wallet
+    /// down to "all of input - fee" without the selector reserving
+    /// min_utxo for a change output that ends up folded into the fee
+    /// anyway.
     #[test]
     fn ada_only_send_can_drain_to_fee() {
         let payment = payment_from_canonical();