codex-cli-architect-mcp
Version:
MCP server for OpenAI Codex CLI - Enable AI-powered coding assistance through Model Context Protocol
38 lines • 2.04 kB
JavaScript
import { z } from 'zod';
import { DEFAULT_MODEL } from '../constants.js';
import { executeCodexCommand } from '../utils/codexExecutor.js';
export const reviewSchema = z.object({
target: z
.string()
.min(1)
.describe('What to review - feature/component (e.g., "user authentication"), or Git context (e.g., "git diff", "staged changes", "changes in feature-branch"). Codex will identify relevant files automatically'),
focus: z
.array(z.enum(['security', 'performance', 'readability', 'best-practices', 'architecture']))
.optional()
.describe('Specific areas to focus on'),
languages: z
.array(z.string())
.optional()
.describe('Programming languages involved (e.g., ["typescript", "python"])'),
model: z.string().optional().describe('AI model to use'),
});
export const codexReviewTool = {
name: 'codex_review',
description: 'Review code for a feature, component, or Git changes. Codex analyzes the relevant implementation or modified files, providing contextual feedback and improvement suggestions',
schema: reviewSchema,
async execute(args) {
const { target, focus = ['security', 'performance', 'readability', 'best-practices', 'architecture'], languages, model = DEFAULT_MODEL, } = args;
let prompt = `Review the following: ${target}\n`;
prompt += `Focus areas: ${focus.join(', ')}\n`;
if (languages && languages.length > 0) {
prompt += `Languages involved: ${languages.join(', ')}\n`;
}
prompt += `\nAnalyze the code and provide actionable feedback with severity levels (high/medium/low) for identified issues. Consider the context and purpose of the changes or feature.`;
const result = await executeCodexCommand(prompt, { model });
if (!result.success) {
throw new Error(result.error || 'Failed to perform code review');
}
return result.output || 'No review feedback generated';
},
};
//# sourceMappingURL=review.tool.js.map