abstract-invariant-generator

Generate abstract invariants for numerical C code using abstract interpretation domains.

1|2|Updated Mar 10, 2026

Search Tags:#verification #formal methods #program analysis #abstract interpretation #invariant generation #numerical code

Ready to run

npx skills add https://github.com/santosomar/general-secure-coding-agent-skills --skill abstract-invariant-generator-santosomar

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: abstract-invariant-generator
Source: https://github.com/santosomar/general-secure-coding-agent-skills/tree/main/skills/verification/abstract-invariant-generator
Command: npx skills add https://github.com/santosomar/general-secure-coding-agent-skills --skill abstract-invariant-generator-santosomar

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps find invariants in numerical code when standard methods fail, particularly for complex relationships between variables.

Core Features & Use Cases

Abstract Interpretation: Leverages various abstract domains (Sign, Interval, Octagon, Polyhedra, Congruence) to over-approximate program states.
Invariant Synthesis: Iteratively computes loop invariants through fixpoint iteration, with widening and narrowing for non-stabilizing loops.
Use Case: Verifying a critical numerical algorithm where simple trace-based invariant inference is insufficient; use Octagon domain to find relational invariants like i + j = n.

Quick Start

Use the abstract-invariant-generator skill to find invariants for the provided C code snippet using the interval domain.

Frequently Asked Questions about abstract-invariant-generator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate loop invariants for numerical code when standard inference fails?▼

Abstract interpretation over-approximates program states using domains like intervals, octagons, and polyhedra. It iteratively computes fixpoints with widening and narrowing to synthesize sound relational invariants for numerical code.

When should I use octagon or polyhedra domains for formal verification?▼

Use octagon or polyhedra domains when verifying multi-variable relationships in numerical code. They capture relational constraints like i + j = n, which simpler interval-based abstract interpretation cannot express.

What is the best way to find relational invariants in C code snippets?▼

Apply abstract interpretation using the octagon or polyhedra domains to C code snippets. This over-approximates program states and computes fixpoints to infer multi-variable relational invariants that trace-based methods miss.

Can abstract interpretation handle non-stabilizing loops in program analysis?▼

Yes, abstract interpretation handles non-stabilizing loops by applying widening and narrowing techniques during fixpoint iteration. This accelerates convergence while maintaining sound invariant generation for program analysis.

Why does simple trace-based invariant inference struggle with multi-variable relationships?▼

Trace-based inference relies on observed executions and cannot over-approximate all reachable program states. Abstract interpretation solves this by computing fixpoints across relational domains like octagons to discover sound multi-variable invariants.