Lattice auditor
Skill Yue-Zhou1/zkcrypto-audit/plugins/post-quantum-auditors/skills/lattice-auditor
Audit lattice-based cryptography for LWE/RLWE parameter soundness, noise sampling correctness, rejection-sampling safety, and decryption-failure assumptions.From its SKILL.md
npx -y skills add Yue-Zhou1/zkcrypto-audit --skill lattice-auditorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.6 KB, 500 tokens by cl100k_base, as published. Nobody here has run it
lattice-auditor
Domain auditor for lattice-based cryptographic implementations.
When to Use
- Auditing LWE/RLWE parameter sets and claimed security levels
- Reviewing noise/distribution sampling and rejection-sampling code
- Checking seed/domain separation and key derivation paths
- Validating decapsulation behavior and decryption-failure assumptions
When NOT to Use
- Traditional ECC-only audits without lattice constructs
- Standardized KEM API/state/serialization/decapsulation conformance
(ML-KEM/FIPS 203 implicit rejection, input checks, rounding) ->
pqc-kem-auditor; this skill keeps generic LWE/RLWE parameter and sampler reasoning - High-level protocol reviews detached from parameter/sampling internals
- Declaring suspected lattice issues confirmed without verification gates
Core Review Areas
- Parameter-set provenance and claimed security margins
- Distribution/noise sampling correctness and bias control
- Domain separation for seeds and derivation contexts
- Decryption failure bounds and constant-time decoding boundaries
- KEM decapsulation checks and reject-path correctness
Workflow
Phase 1: Parameter and provenance intake
- Read
references/lattice-checklist.md - Execute
workflows/parameter-review.md - Map claimed scheme parameters to concrete code constants
Phase 2: Sampling and failure-path review
- Trace sampling paths for noise and rejection behavior
- Validate decryption failure assumptions against implementation behavior
- Confirm decapsulation reject paths are strict and constant-time where required
Phase 3: Pattern hunt
- Read
references/finding-patterns.md - Prioritize sampling bias, seed reuse, and failure-bound underestimation
Phase 4: Handoff
- Send surviving findings to
crypto-fp-check - Use
zkbugs-indexonly after the finding survives verification
Output Contract
Produce a lattice-specific handoff that includes:
- The parameter set, sampling routine, or decapsulation path involved
- The concrete failure bound or bias condition at risk
- Whether the issue is parameter, sampling, seed-separation, or reject-path related
- The next verification or reporting route
Reference Index
What ships with it: 4 files
3.4 KB alongside SKILL.md
agents/
- openai.yaml422 B
references/
- finding-patterns.md692 B
- lattice-checklist.md1.2 KB
workflows/
- parameter-review.md1.1 KB