agentsclimarketplace

Side channel auditor

Skill Yue-Zhou1/zkcrypto-audit/plugins/implementation-safety/skills/side-channel-auditor

Audit timing, cache, memory-access, and power-analysis leakage patterns, including compiler and feature-flag regressions that break constant-time assumptions.From its SKILL.md

Install
npx -y skills add Yue-Zhou1/zkcrypto-audit --skill side-channel-auditor

Assembled 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

2.9 KB, 580 tokens by cl100k_base, as published. Nobody here has run it

side-channel-auditor

Domain auditor for constant-time assumptions and microarchitectural leakage risk.

When to Use

  • Reviewing secret-dependent control-flow and memory-access behavior
  • Auditing cache, branch-predictor, and power-related leakage surfaces
  • Checking compiler/feature-flag combinations for constant-time regressions
  • Verifying zeroization and error-handling latency do not leak secrets

When NOT to Use

  • Protocol-level correctness audits without side-channel threat surfaces
  • Purely functional bugs unrelated to timing/cache/power leakage
  • Active fault-injection attacks where the attacker glitches/corrupts the computation (RSA-CRT Bellcore, DFA, verification-skip) -> fault-injection-auditor; this skill covers PASSIVE leakage only
  • Declaring side-channel concerns confirmed without evidence-backed verification

Core Review Areas

  1. Secret-dependent branches and control-flow decisions
  2. Table lookups and cache-line dependent access patterns
  3. Variable-time arithmetic and modular-reduction hotspots
  4. Compiler/feature-flag regressions that reintroduce leakage
  5. Operational side effects (logging, errors, zeroization) that leak validity

Workflow

Phase 1: Secret-bearing inventory

  • Read references/side-channel-checklist.md
  • Execute workflows/constant-time-review.md
  • Enumerate secret-bearing values and trace their influence on control/data flow

Phase 2: Constant-time path analysis

  • Validate constant-time helper usage and fallback behavior
  • Inspect branch and memory-access patterns under secret variation
  • Review compiler settings and feature flags for regression risk

Phase 3: Pattern hunt

  • Read references/finding-patterns.md
  • If the codebase includes a ZK prover, read references/zk-prover-patterns.md before completing the pattern hunt
  • Prioritize secret-branch, table-index, optimization, and error-latency findings

Phase 4: Handoff

  • Send surviving findings to crypto-fp-check
  • Use zkbugs-index only after the finding survives verification

Output Contract

Produce a side-channel-specific handoff that includes:

  • The secret-bearing value and affected control/data path
  • The leakage class (timing, cache, memory, power, or operational)
  • Tool-backed evidence status (dudect, ctgrind, microbenchmarks) if available
  • The next verification or reporting route

Reference Index

What ships with it: 5 files

5.5 KB alongside SKILL.md

agents/

workflows/

Keep looking

Skills are one crate of 326,852. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.