eiffel-review

Automate Phase 2 adversarial review cycles for Eiffel contracts with human approval gating.

Updated Jan 22, 2026
One-click install
npx skills add https://github.com/simple-eiffel/eiffel-spec-kit --skill eiffel-review
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: eiffel-review
Source: https://github.com/simple-eiffel/eiffel-spec-kit/tree/main/eiffel-review
Command: npx skills add https://github.com/simple-eiffel/eiffel-spec-kit --skill eiffel-review

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates the Phase 2 adversarial review workflow for Eiffel contracts, enabling a structured, multi-agent assessment while enforcing a formal human gate before proceeding.

Core Features & Use Cases

  • Progressive AI chain: Orchestrates Ollama → Claude → Grok → Gemini → Human reviews to surface risks and improvements in Eiffel contracts.
  • Artifact generation: Produces implementation planning artifacts like approach.md and synopsis.md.
  • Governance & gating: BLOCKS until explicit human approval, ensuring compliance with review policies.
  • Use Case: A team analyzing a new Eiffel library can run /eiffel.review to generate a centralized review trail and a clear set of action items.

Quick Start

Run the /eiffel.review command against your Eiffel project path to initiate the phase-2 adversarial review cycle.

Frequently Asked Questions about eiffel-review

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

FAQPage Schema
How do I automate an adversarial review for Eiffel contracts?

You automate an adversarial review for Eiffel contracts by running the /eiffel.review command, which orchestrates a multi-AI chain to generate review artifacts and block until human approval.

What is the Phase 2 review process for Eiffel libraries?

The Phase 2 review process for Eiffel libraries is a structured assessment cycle that uses multiple AI models to surface risks, generates planning artifacts, and enforces a human gate before proceeding.

How do I generate approach.md and synopsis.md for an Eiffel project?

You generate approach.md and synopsis.md for an Eiffel project by initiating the phase-2 adversarial review cycle, which automatically produces these implementation planning artifacts during the AI chain execution.

Can I use multiple AI models like Claude and Gemini for contract review?

Yes, you can use multiple AI models for contract review as the workflow orchestrates a progressive chain across Ollama, Claude, Grok, and Gemini before requiring explicit human approval.

Does the Eiffel contract review workflow block without human approval?

Yes, the Eiffel contract review workflow blocks without human approval, enforcing a formal governance gate that ensures compliance with review policies before any further processing proceeds.

How is review evidence traced in an Eiffel workflow?

Review evidence is traced in an Eiffel workflow by organizing prompts and generated artifacts under the .eiffel-workflow directory, creating a centralized trail for compliance and auditability.