Peer Reviewer Skill

AI-powered academic rigor for your research.

Peer Reviewer is an OpenClaw skill that simulates a multi-agent academic peer review process. It employs a "Council of Agents" to deconstruct arguments, find contradictions in established literature, and render a final judgment on the merit of a scientific paper or claim.

🧠 How It Works

The system uses three specialized AI agents:

1. The Deconstructor: Parses raw text into a formal Logic Graph (Toulmin Model), extracting claims, premises, and evidence without judgment.
2. The Devil's Advocate: Takes the claims and actively searches for contradictions (using Google Serper & ArXiv). It looks for theoretical conflicts, empirical contradictions, and prior art.
3. The Judge: Evaluates the Logic Graph against the Devil's Advocate's objections. It scores the paper on logical coherence, foundational integrity, and empirical falsifiability.

🚀 Installation

npm install @sschepis/peer-reviewer

🛠️ Usage

As a Library (Dependency Injection)

The core PeerReviewer class is decoupled from specific LLM or Search implementations. You can provide your own adapters or use the included ones.

import { PeerReviewer, OpenClawLLMAdapter, SkillSearchAdapter } from '@sschepis/peer-reviewer';

// Use standard OpenClaw/Env variables (OPENAI_API_KEY, GEMINI_API_KEY)
const llm = new OpenClawLLMAdapter();

// Use the Serper CLI skill (interop)
const search = new SkillSearchAdapter('serper-tool');

const reviewer = new PeerReviewer(llm, search);
const report = await reviewer.review("My scientific claim...");

Standalone CLI

You can run the reviewer directly. It will automatically detect available providers (Google Cloud, OpenAI, Serper).

# Review a file
node dist/index.js "/path/to/paper.txt"

⚙️ Configuration

The tool adapts to your environment:

• LLM: Checks OPENAI_API_KEY, GEMINI_API_KEY, or GOOGLE_APPLICATION_CREDENTIALS.
• Search: Checks for serper-tool (CLI skill), SERPER_API_KEY, or falls back to ArXiv (free).

License

MIT