agentleFS
odinfreeodinfree/storm-ecdsa-harness/skills/quantum-replay-arithmetic/SKILL.md

quantum-replay-arithmetic

Skill4 starsChanged 21 days agoMITView on GitHub

Co-design reversible GCD walks, transcript encodings and payload replay for structural point-addition gains; compare full arithmetic families and cleanup.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the shared contract before a trial.

    line 10

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Widerchäuer (quantum-replay-arithmetic)

Change the walk, the information it records, or the way its consumer reconstructs arithmetic. Read both arithmetic papers.

Read the shared contract before a trial. Start with the five-line post; the shared native-test order, stop rules and promotion gate apply.

Practical lane

Audit the active dispatcher before designing: list recurrence, transcript alphabet, packing/decoder, coefficient or payload updates, register alternation, square/reduction and both divisions. Mark each proposed feature already present, absent or uncertain. Reintroducing an inherited feature is not a contribution.

Compare coherent families rather than mixing incompatible fragments: compressed multi-step history, fixed-alternation history, and in-place multiplication/division with delayed coefficient reconstruction. Preserve a common input/output and cleanup contract. A classical-addend circuit and a quantum-addressed window circuit have different interfaces and costs.

First experiment: build a small independent arithmetic model for one transition plus reverse replay. Test complete histories for small odd moduli and supported inputs, including zero payload, terminal states and signed/canonical boundaries. For the chosen family, specify how full-width convergence and register capacity are justified.

Then test one complete numerator/denominator boundary before replacing the point adder. Price the decoder, packed history, both arithmetic directions, payload corrections and temporary products. If comparing fused replay, isolate its cost using an otherwise identical family; do not credit unrelated square or lookup changes to the recurrence.

Structural extension

Design the transcript for the exact next consumer rather than reconstructing generic decisions. Consider eliminating full-width comparisons or materialized coefficients. A larger transcript can win if it removes more expensive work; measure the complete tradeoff.

Retain the papers' structured counterexamples as regressions. Fresh random passes cannot erase known failures. Deliver recurrence/inverse specifications, transcript occupancy, full-width regressions and whole point-add resources. Falsify on noninvertible state transitions, unbounded history, unsupported exceptional inputs or unfavorable full-path economics.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.