Proposal: Proof of Audits - Continuous Security & Privacy-Preserving Trust Layer for Canton

CLOSEDPull Request

by Proofofaudit21-08-2026Declined

daml-toolingneeds-champion

Hi Canton team - I am MK, solo founder of Proof of Audits (https://proofofaudits.com).

I built a contest platform where the report does not die after the audit. We run T4->T1: Function -> Contract -> Cross-Contract -> System in sequence. T3 cannot start until T4 is validated, T2 until T3, T1 until T2. No duplicate pay, no blind spots.

For Canton I want to ship this T4->T1 engine as open-source, all at once:

Give it a .dar, it lists every place package A can use package B's template/choice (file, line, versions, choice, consuming) - JSON + graph for CI
Permissioned Trust Passport: public green check + /1400, private vault for full bundle/interview (banks don't leak IP) + ZK attestation
Contract Shield for Canton: free extension checks "deployed DAR == reviewed DAR" + signatories before you sign

This fills the gap outside Zenith (EVM) and Certora (analyzer) - guaranteed coverage.

Full proposal is in the file below - pasted directly here as you do for other PRs:

---

Development Fund Proposal - Proof of Audits for Canton

---

Hi Canton team - I am MK, I built Proof of Audits.

What I built: I run a contest platform where the report does not die after the audit. We find bugs with contests, then we keep the proof alive - we match the deployed DAR to the reviewed DAR, we check authority controls, we score it /1400, and we show it in a Trust Passport. Anyone can verify before they sign via our free Contract Shield extension.

Our core is T4->T1. I do not run a crowd where 10 people check the same lines. I run 4 tiers in sequence:

T4 checks functions (edge cases)
T3 checks contracts (must first validate T4)
T2 checks cross-contract calls (must first validate T3)
T1 checks the whole system (must first validate T2)

No duplicate pay, no blind spots. T3 cannot start until T4 is done, and so on. That is our main.

Why this matters for Canton: Canton hides data by default - only the right parties see the right data. That is great for banks, but it also hides the exact risk: in Daml, the bug is not reentrancy, it is "package A quietly exercising a choice of package B and moving money" or a bad signatory. That risk lives in T2 and T1 - the layers most audits skip. Canton has EVM support (Zenith) and a great Daml analyzer (Certora), but no one guarantees T4->T1 was actually covered. I want to build that guarantee for Canton.

What I will ship for Canton - all at once, not piecemeal:

I will ship one unified system - CLI + Passport + Shield - built together:

T4-T1 Daml engine: Give it a .dar, it builds the call graph, it checks all 4 tiers in order. T2 specifically lists every place where one package can use another package's template/choice - with file, line, package versions, template/choice name, and whether it consumes. You get JSON you can use in CI plus a visual graph.

Permissioned Trust Passport: Because banks cannot leak IP, I will make passports permissioned. Public sees a green check and a /1400 score. The full bundle and the interview stay in a private vault - only a KYC'd counterparty you approve can open it. I can also give a ZK attestation: "keys are verified" without revealing who holds them.

Contract Shield for Canton: My free extension already does this for EVM at proofofaudits.com/verify-contract. I will add Canton - it checks "does the DAR on the validator equal the DAR we reviewed?" and "are the signatories right?" before you sign.

All of this runs on the compiled .dar - the same file you deploy. It only reads, it never changes the ledger. It fits right into daml build / dpm / CI. I will open-source it Apache-2.0.

---

How I will deliver it - 4 months, milestone-based

M1 - Build the T4->T1 engine (2 months, 210k CC): I will deliver a CLI that runs on Linux/Mac/Windows, takes a .dar, runs T4->T3->T2->T1 in order, and blocks T3 if T4 has not passed. T2 will report all 6 fields for every cross-package use. I will docs it and release it Apache-2.0, tested on 2 real Daml projects.

M2 - Pilot it on real Canton apps (1 month, 105k CC): With your intros, I will run it on Splice + 5 voting member apps. I will issue their first permissioned passports and fix what you tell me to fix.

M3 - Shield + broad adoption (1 month, 105k CC): I will ship the Canton Shield extension, verify DAR hash + authority on a live validator, run the engine on 5 more production apps, and publish a CI template + short video.

Total: 420k CC, paid per milestone per CIP-0100 (0.15 USD ref, 30-day MA band 33.3%).

---

How you can check I delivered

M1: You run prove --dar app.dar and you get JSON + graph with the 6 fields, and you see T3 blocked until T4 passes. 3 voting members try it and say "yes".

M2: Splice + 5 members ran it, got passports, and say they want to keep using it.

M3: Shield shows green for a correct DAR and red for a wrong DAR on a validator. 5 more members ran it and endorsed. CI template works.

---

Why I am asking Canton

You give me milestone funding, I give Canton open infra that makes every future audit provably covered. This fills the gap between Zenith (EVM) and Certora (analyzer) - I give you the guaranteed-coverage model.

I will also write a joint blog on cross-package delegation risk and run a workshop for Canton devs.

About me: Solo founder, Top 10 Sherlock, 180+ repos, Cyfrin certified. I live at proofofaudits.com (beta 30 Jul 2026, pre-revenue, 40 auditors scored). I will maintain this for 12 months.

Happy to jump on a 15-min call and adapt this to what SIG daml-tooling needs most.

License: Proposal CC0-1.0, Code Apache-2.0