BENJI — Verification Runbook

Instrument: Franklin OnChain U.S. Government Money Fund (FOBXX) Issuer: Franklin Templeton Asset class: Treasury (tokenized money market fund) Last verified: 2026-06-02 Status: Live, multi-chain (Stellar primary + Ethereum), composite score 41

This is the canonical record of how we verified BENJI. It captures address provenance, scoring evidence sources, gap attribution per dimension, and what's pending. It is the "show your work" document for this instrument. Update on every methodology change or new verification action.

This is also the template for every other instrument's runbook. Copy and adapt for usdy.md, ousg.md, buidl.md, usyc.md, paxg.md, xaut.md.

1. Address provenance — how we know which contracts are real

BENJI exists on 8 chains. Each chain has its own contract or mint address, and several chains have scam look-alike tokens that share the BENJI ticker. The address is the load-bearing identifier, not the ticker.

Stellar (primary venue, ~95% of supply)

• Asset code: BENJI
• Issuer address: GBHNGLLIE3KWGKCHIKMHJ5HVZHYIK7WTBE4QF5PLAKL4CJGSEU7HZIW5
• Source: Franklin Templeton Benji DevHub — https://digitalassets.franklintempleton.com/benji/benji-contracts/
• Verification steps (verbatim, 2026-06-02):

Ethereum

• Contract: 0x3DDc84940Ab509C11B20B76B466933f40b750dc9
• Source: Franklin Benji DevHub (same URL above) and Etherscan token tracker
• Verification: Etherscan-verified contract, deployer addresses traceable to Franklin Templeton operational keys
• Confidence: ≥98%

Base, Polygon, Arbitrum, Avalanche (catalog'd, currently inactive)

• All four addresses from Franklin's Benji DevHub (single canonical source maintained by the issuer)
• Each address verified against the corresponding chain's block explorer for deployer + token name
• Confidence: ≥95% per chain
• Active = false in lib/instruments.ts and instrument_deployments.active = false in the DB. They will become active once the L2 transfer collector ships (see docs/multi-chain-design.md and "Pending" section below).

Solana (catalog'd, deferred)

• Mint: 5Tu84fKBpe9vfXeotjvfvWdWbAjy3hqsExvuHgFqFxA1
• Source: Franklin Benji DevHub
• Verification: Confirmed via Franklin's own developer documentation. A Pump.fun look-alike token (`76B7YRDaL1zeVSVzxB4RmrQJQCFK1YVxtJDn7p89pump`) exists on Solana and would be a misidentification if relying on Phantom wallet's autocomplete or Solscan ticker search. Mint address from official source is the only safe identifier.
• Confidence: ≥95% on address; deferred from runtime until Solana collector ships.

Aptos (catalog'd, deferred)

• Address: 0x7b5e9cac3433e9202f28527f707c89e1e47b19de2c33e4db9521a63ad219b739
• Source: Franklin Benji DevHub
• Verification: Source identification only; no Aptos collector planned in current launch scope.

2. Scoring evidence — where every dimension's facts came from

Yield Accuracy (dimension scores up to 25)

• Approach for BENJI: peg_nav — instrument targets $1.00 NAV; yield is paid via distribution events (new BENJI minted), not price appreciation
• Source for published rate: SEC EDGAR CIK 0001786958 (Franklin Templeton Trust series S000067043 = FOBXX). Monthly N-MFP3 filings include 7-day SEC yield, AUM, WAM/WAL, stable price per share
• Fetcher: lib/jobs/collectAttestations.ts::fetchFranklinNmfp + lib/edgar.ts
• Last fetch: 2026-05-31 (6 filings ingested, 100% match rate to S000067043)
• Score component: moderate — Yield Accuracy gets 5/25 because rate is monthly-cadence (not daily) and BENJI's distribution model means delta-vs-published is structurally near-zero (the rate IS the distribution), but we can't measure it on-chain

Liquidity (dimension scores up to 25)

• Approach: market liquidity in DEX pools relative to outstanding supply
• Source: raw_pool_state for Ethereum BENJI; no equivalent on Stellar (BENJI on Stellar is a regulated fund — transfers happen via Franklin's authorized chains, not DEXs)
• Score component: low — BENJI is not a DEX-traded asset by design. This is structurally correct; it's not a "gap" so much as a property of the instrument type.

Transparency (dimension scores up to 25)

This is where most of BENJI's score loss comes from, and it's the most important dimension for the gap-attribution narrative.

• attestationFreshness: Franklin publishes monthly N-MFP3 filings via SEC EDGAR. Score: moderate (monthly cadence, not daily). Gap: Franklin could publish a machine-readable daily attestation feed at a stable URL (similar to what OUSG does); they don't.
• disclosureConsistency: N-MFP3 cadence is reliable. Score: moderate.
• regulatoryStanding: SEC-registered money market fund (1940 Act). Score: high.
• governanceQuality: institutional. Score: high.

Concentration (dimension scores up to 25)

• Source for top holders: raw_holder_snapshots on Ethereum; we don't yet have Stellar transfer events or holder snapshots, so concentration is measured only on the Ethereum slice
• Score component: Stellar deployment scored 39 vs Ethereum 41 — the 2-point difference is the concentration dimension lacking Stellar holder data. This is a temporary_collector_gap (engineering — the Stellar holder collector is on the roadmap), not an issuer gap.

3. Gap attribution per dimension

This is the heart of Pivot 7. Each dimension's points-loss is categorized and the path to closing it is named.

The single most important issuer-side gap: Franklin publishing daily machine-readable attestation. OUSG does it; BENJI doesn't. Same asset class, same regulatory bracket. If closed, BENJI's composite jumps roughly 41 → 53.

4. Multi-chain coverage status

Total BENJI supply across all venues (per one-time multi-chain read 2026-06-02): ~$795M — vs the $571M currently in the published Index. The $224M difference is the gap closing once L2 + Solana collectors ship.

5. Pending items

• Daily Franklin attestation feed. Open question for the issuer. We should request via FTDigitalAssets@franklintempleton.com — track response in verification_log table (Pivot 7 Option A, post-launch).
• L2 transfer collector. Engineering. Once shipped, flip BENJI Base/Polygon/Arbitrum/Avalanche to active. Adds ~$177M to BENJI's tracked supply.
• Solana collector. Engineering. Adds ~$648M (the largest single increment).
• Stellar holder collector. Engineering. Closes the 2-point concentration delta between Ethereum and Stellar.
• Distribution-mode yield rubric. Methodology revision (v2.x). Allow distribution-based yield to be scored via mint-event analysis instead of price-vs-rate delta. Would help BENJI, BUIDL, and any future distribution-yield instrument.

6. Change log

• 2026-06-02: First verification pass. Stellar collector landed. BENJI multi-chain registry populated; L2 deployments catalog'd as inactive. Initial gap attribution captured.