Prove or refute code correctness claims using induction, contradiction, and counterexamples.
npx skills add https://github.com/NlightNFotis/skills --skill proof-tactics
Please help me install this Agent Skill. Skill: proof-tactics Source: https://github.com/NlightNFotis/skills/tree/main/proof-tactics Command: npx skills add https://github.com/NlightNFotis/skills --skill proof-tactics
This Skill helps you justify that code is correct for all relevant inputs, not just the handful of cases you tested. It is especially useful when loops, recursion, exhaustive branching, or universal claims about behavior make informal reasoning too weak.
Use the proof-tactics skill to verify that a recursive function terminates and returns the correct result for all finite inputs.
High-intent search queries and answers about installing and using this skill.
Prove code correctness by applying mathematical proof tactics like induction, contradiction, and case analysis to verify behavior across all inputs, not just tested cases. This requires precise claim statements and explicit loop invariants.
Verify loop invariants and termination by defining explicit invariants, showing initialization and maintenance, and providing well-founded decreasing measures for recursion. This connects loop termination directly to the desired postcondition.
Find counterexamples for exhaustive branching by systematically searching adversarial examples, empty inputs, singleton cases, boundaries, duplicates, and numeric extremes before trusting the universal behavior claims.
You need formal reasoning when loops, recursion, exhaustive branching, or universal behavior claims make informal testing too weak to justify correctness across all relevant inputs and execution paths.
Yes, you can validate behavior-preserving refactors by applying proof tactics like contrapositive and case analysis to exhaustively verify that the refactored logic maintains identical behavior across all inputs and state transitions.
ShiienConvert informal mathematical claims into precise LaTeX statements with assumptions and notation.
rainoftimeGenerate ACSL formal annotations for C/C++ programs to verify with Frama-C.
mashharukiValidate smart contracts with Foundry fuzz, fork, and invariant testing.
Leo1349Generate rigorous mathematical proofs and validate claims in machine learning theory.
