Skip to content

Solution: LP-0002 Private M-of-N Multisig #92

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Tranquil-Flow wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Solution: LP-0002 Private M-of-N Multisig #92

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
94 changes: 94 additions & 0 deletions solutions/LP-0002.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,94 @@
# LP-0002 Solution: Private M-of-N Multisig

**Submitted by:** Tranquil-Flow

## Summary

This submission implements a private M-of-N multisig primitive for LEZ. Shielded members approve a proposal client-side, produce unlinkable per-proposal nullifiers, and execute a threshold-gated action without revealing which members approved. The repository includes the Rust workspace, SDK, SPEL IDL, RISC0 heavy-lane artifacts, public LEZ testnet evidence, a native Qt/QML Basecamp `.lgx` package with alice/bob install evidence, and reproducible evaluator scripts. The fresh human-recorded narrated demo is available at https://youtu.be/Wssfp_rkC54.

## Repository

- **Repo:** https://github.com/Tranquil-Flow/lp-0002-private-multisig
- **License:** MIT
- **Narrated demo:** https://youtu.be/Wssfp_rkC54

## Approach

The system is split into audit-friendly layers:

1. `core/` implements member commitments, proposal binding, threshold relation checks, context-bound nullifiers, replay identifiers, and the public journal privacy boundary.
2. `sdk/` exposes a high-level `MultisigSession` workflow for create, propose, approve, prove, and execute.
3. `consumer-demo/` is a standalone clone-and-run integration app that imports the public crates as library dependencies and exercises five realistic multisig scenarios.
4. `methods/` and `methods/guest/` contain the RISC0 threshold-proof guest and executable wrapper guest.
5. `host/` verifies real RISC0 receipt artifacts, prepares LEZ/NSSA payloads, and records file-backed evidence for the heavy lane.
6. `lez-program/` models the LEZ verifier gate with deterministic account-state mutation and replay protection.
7. `basecamp-app/` contains the native Qt/QML Logos Basecamp plugin package; `flake.nix`/`flake.lock` package it as `.lgx`, and `submission/BASECAMP_INSTALL_EVIDENCE.json` records M4 alice/bob profile install evidence.

The proof design keeps member identity private. The public journal reveals multisig/proposal/action binding, threshold and approval counts, sorted nullifiers, member root/count, and proof identifiers, but not raw member secrets or member commitments. Nullifiers bind approvals to the proposal context so a member cannot approve twice while remaining unlinkable across proposals.

The LEZ evidence is intentionally explicit about the current transport boundary. The full RISC0 receipt is verified host-side and persisted as file-backed evidence. The included LEZ public-testnet wrapper transaction carries compact receipt/journal commitments because raw receipt bytes exceed the current public-program session transport limit. The reproducible `verify_and_execute_bytes` wrapper program is deployed and executed on the public LEZ testnet (https://testnet.lez.logos.co/): deploy tx `82516880f60c2076d78b28ad7b147ac0b05ed247b7bc33a27ac8f68b1d809c56` in block `39547` and execute tx `cb8bfd5afca3c88a99b12b42a6875bcc2cad419d394da0e39d8ca463ee376697` in block `39548`, signed by funded testnet account `6iArKUXxhUJqS7kCaPNhwMWt3ro71PDyBj7jwAyE2VQV`. The confirmed transaction evidence is recorded in `submission/TESTNET_EVIDENCE.json`.

## Success Criteria Checklist

### Functionality

- [x] Any M-of-N member holding a shielded LEZ account can submit an approval without revealing their identity to on-chain observers or other members.
- [x] The verifier confirms a threshold of M approvals without recording which members approved.
- [x] Double-vote prevention is implemented through context-bound nullifiers.
- [x] A completed execution is unlinkable to an individual member's shielded account.
- [x] Proof generation runs client-side; the repository includes real `RISC0_DEV_MODE=0` proof artifacts and host verification scripts.
- [x] A reference threshold-gated action is delivered through the SDK, consumer demo, LEZ-shaped wrapper, and recorded public-testnet evidence.
- [x] At least one multisig instance/proposal/approval/execution path is evidenced in `submission/TESTNET_EVIDENCE.json` with confirmed transaction hash and block reference.
- [x] Full documentation and a clean public repository are delivered.

### Usability

- [x] Module/SDK provided via the `sdk/` crate and `MultisigSession` API.
- [x] Logos Basecamp GUI package provided under `basecamp-app/`, including native Qt/QML plugin source, metadata, CMake build instructions, `.lgx` packaging, and alice/bob profile install evidence.
- [x] SPEL IDL provided under `interfaces/lp0002.idl.json` with discriminators and documented instruction/error surfaces.

### Reliability

- [x] Proof generation and verification failures surface deterministic typed errors.
- [x] Partial approvals are preserved and resumable through serde round-tripping of the approval accumulator.
- [x] Verifier errors and SDK errors are deterministic and documented in `docs/PROTOCOL.md`.

### Performance

- [x] Proof-generation measurements, receipt/journal sizes, wrapper payload metrics, and public LEZ testnet cost evidence are documented in `submission/BENCHMARKS.md` and `submission/LEZ_COST_BENCHMARKS.json`.
- [x] Current LEZ tooling does not expose stable per-transaction CU counters; the submission records this limitation explicitly rather than inventing CU numbers.

### Supportability

- [x] Program deployment and execution evidence is recorded for the public LEZ testnet (https://testnet.lez.logos.co/).
- [x] End-to-end LEZ/RISC0 smoke scripts are included, with evaluator commands in `submission/EVALUATOR_GUIDE.md`.
- [x] CI configuration is present in the linked repository as `.github/workflows/ci.yml` and `.gitlab-ci.yml`; local/evaluator validation evidence is also documented in `submission/CI_EVIDENCE.md`.
- [x] README documents deployment, program identifiers, SDK usage, CLI/demo usage, and Basecamp package usage.
- [x] A reproducible demo script is provided at repository root as `demo.sh`; the heavy-lane path is available through `scripts/demo-heavy-lane.sh`.
- [x] Fresh narrated demo video: https://youtu.be/Wssfp_rkC54

## FURPS Self-Assessment

### Functionality

The workspace implements the private threshold relation, proposal binding, nullifier-based double-vote prevention, replay protection, SDK integration, consumer app, native Basecamp package, SPEL IDL, real RISC0 proof verification, and confirmed public LEZ testnet inclusion evidence. The main caveat is transparently documented: compact commitments are transported through the wrapper transaction while the full receipt remains host-verified and file-backed due to current LEZ session limits.

### Usability

Evaluators can start with `./demo.sh` for the fast clone-and-run consumer flow, then use `submission/EVALUATOR_GUIDE.md` for the heavy-lane evidence path. Developers can import the SDK crate, inspect the consumer demo, or build the native Basecamp plugin from `basecamp-app/`.

### Reliability

The Rust workspace includes unit and integration tests for privacy boundaries, duplicate approvals, invalid indices, insufficient approvals, replay protection, serialization/resume behavior, IDL consistency, and verifier errors. Readiness validators catch stale documentation, missing artifacts, IDL drift, and publication-gate regressions.

### Performance

The submission includes safe-lane timings, RISC0 fixture measurements, serialized receipt/journal sizes, NSSA wrapper payload metrics, account counts, and public-testnet inclusion evidence. LEZ per-transaction CU counters are not currently exposed by the available tooling, so the cost evidence records that limitation in machine-readable form.

### Supportability

The repository is MIT licensed, public, documented, and organized as a reproducible Rust workspace. `submission/EVALUATOR_GUIDE.md`, `docs/SPEC_COMPLIANCE.md`, `docs/PROTOCOL.md`, `submission/TESTNET_EVIDENCE.json`, and `submission/BENCHMARKS.md` give reviewers the paths, commands, evidence identifiers, and known limitations needed to independently assess the submission.

## Terms & Conditions

I have read and agree to the Logos Lambda Prize TERMS.md and confirm that this submission is original work published under the MIT license.
Loading