What is a claim tree in the Rationale skill?

A claim tree is a hierarchical structure that decomposes the broad goal of 'code adequacy' into specific, verifiable subclaims such as structural correctness, behavioral logic, and non-functional requirements.

When should I use the /rationale command?

Use it when you need a high level of assurance that a function meets its requirements, such as during a security audit, a complex refactor, or when implementing critical business logic.

How does Rationale handle formal verification?

It identifies claims that can be proven mathematically, drafts candidate Dafny specifications (preconditions and postconditions), and offers to run them through a formal verifier.

What is the difference between behavioral and static verification in this skill?

Behavioral verification involves generating and running tests to check runtime logic, while static verification involves citing specific evidence within the source code, such as type annotations or variable assignments.

Rationale: Code Adequacy Verification

Name: Rationale: Code Adequacy Verification
Author: nicholls-inc

bynicholls-inc

•

Security & Testing

Builds hierarchical claim trees to verify that code satisfies requirements through formal, behavioral, and static analysis.

The Rationale skill bridges the gap between formal verification and semi-formal reasoning by constructing a structured argument for code adequacy. It decomposes high-level requirements into a tree of verifiable subclaims—categorized into formal proofs, behavioral tests, static evidence, or human judgment. By generating candidate Dafny specifications and property-based tests, it provides a traceable, transparent checklist that ensures every aspect of a module's behavior is accounted for and verified against its original requirements.

Key Features

01Automated edge-case and property-based test generation

02Hierarchical claim tree construction

03Dafny specification generation and iteration

04Traceable verification checklists with evidence citations

05Multi-modal verification (Formal, Behavioral, Static, Semantic)

063 GitHub stars

Use Cases

01Proving the correctness of mission-critical algorithms

02Generating comprehensive verification documentation for regulatory audits

03Decomposing complex business requirements into verifiable code paths

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nicholls-inc/claude-code-marketplace rationale

For use in Claude.ai and ChatGPT

Key Features

01Automated edge-case and property-based test generation

02Hierarchical claim tree construction

03Dafny specification generation and iteration

04Traceable verification checklists with evidence citations

05Multi-modal verification (Formal, Behavioral, Static, Semantic)

063 GitHub stars

Use Cases

01Proving the correctness of mission-critical algorithms

02Generating comprehensive verification documentation for regulatory audits

03Decomposing complex business requirements into verifiable code paths

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nicholls-inc/claude-code-marketplace rationale

For use in Claude.ai and ChatGPT